Stories
Slash Boxes
Comments

SoylentNews is people

posted by n1 on Monday July 13 2015, @12:42PM   Printer-friendly
from the more-you-share-the-more-you-care dept.

Julien Voisin blogs:

Today, I updated my Firefox, and had a new icon on my toolbar: pocket. I took at quick look at the ToS and privacy policy; here is my tl;dr:

Read it Later, Inc. is collecting a lot of intimate information and is tracking you.

When you share something through Pocket with a friend, the emails contains spying material using malware-like techniques to track your friends.

They are sharing those information with trusted third parties (Could be anyone they are doing business with.).

The policy might change, and it's your responsibility to check Pocket's website to see if it has.

[...] The Pocket implementation is not an extension (while it was available as an extension), it's implemented in Firefox. You can not remove it, only disable it, by going in about:config, since this option is not available in the preferences menu.

What the hell is pocket? on Mozilla's site:

The Pocket for Firefox button lets you save web pages and videos to Pocket in just one click. Pocket strips away clutter and saves the page in a clean, distraction-free view and lets you access them on the go through the Pocket app. All you need is a free account, an Internet connection and the Pocket button.


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2) by Hairyfeet on Sunday July 19 2015, @03:56PM

    by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Sunday July 19 2015, @03:56PM (#211094) Journal

    If you wanna get one of those? I'd suggest just buying whichever is cheapest but I warn ya....IMHO the new ones are about as useful as throwing darts at a dartboard when it comes to telling you what is actually wrong with a board. I went through 3 different ones about 5 years ago trying to find one that actually worked before being told by an engineer friend of mine to give it up, if the board is too fucked to give you the standard beep codes it was beyond fixing. Every one I tried had the same two problems which were 1.- No manual or a manual in bad Chingrish with no real explanation as to "code equals problem" and 2.- They always seem to default to a single code that usually is provably wrong. What I mean by that is they will often give you a code of "bad BIOS" yet you can reset the CMOS and will get the same error code, even if the board will post! I'd also had them spit out an error code on a perfectly functioning board I used to test the thing, giving me codes from bad BIOS to damaged RAM when everything is working beautifully.

    So if you wanna get one to play with? Get this one [amazon.com] but like all of the new ones you'll have to go to the BIOS makers website to find out what code means what for each individual board you test. But in my experience either the board will beep, in which case you can simply look up the error code for that BIOS manufacturer yourself, or it won't in which case the board is too badly damaged to be salvaged by a layman. Nearly every bad board at the shop can be traced by to one of two failures, which are bad traces (which you just aren't gonna fix) or blown caps, which you should be able to see by eye and if you have the time and patience to replace blown caps good luck, never had steady enough hands and keen enough eyes to do solder work myself. With boards being able to be found so cheap off of places like Craigslist and eBay I find its really not worth spending too much time with a bad board, especially considering that if you get an iffy board to work it can just as easily turn around and cook,, often taking any RAM and CPU installed along with it.

    But don't be afraid of any AMD board AM2 or better, while the socket A and 754 boards you had to watch the temps by the time socket AM2 came out all the bugs had been worked out and with a few exceptions (cough foxconn and biostar cough) most AM2/AM2+ boards you are gonna find are gonna be pretty reliable if nothing fancy. I personally love when one of those boards lands in my lap because as I showed you there are a ton of good chips that will fit them that are dirt cheap, it reminds me of the mid 00s with the LGA775 boards where you could fit anything from a netburst celery to a C2Q in 'em.

    Oh and as for PSUs? I like Lepa myself for my gaming builds as well as my own PC, they are modular, come with a rubber jacket that goes around the edges of the PSU to damp vibration, really top notch. For the cheap builds I like Apevia models like this one [amazon.com]. Now you may scoff at old Hairy but here is a little "trick" when it comes to the cheaper PSUs...if you make sure the max the system can pull is only around half what the unit is rated for? The cheap PSUs will often last as long as the good PSU. The reason being that most of those PSUs are full of shit when it comes to its output...think that Apevia is 700w? Bullshit its 550w max if that. So if you were to slap it in a system that could never draw more than 400w-450w? It still has plenty of headroom and its pulling right at its maximum that blows those cheapie PSUs. this rule works whether its Apevia or Rosewill or HongFong fakery, just look at how the unit is put together and figure it'll be around half plus 50w-75w what its rated for and you're good to go. Sure it'd be nice if I could just slap a Lepa bronze in everything but if a used unit is only gonna get $200? it makes no sense putting a $85+ power supply in the thing.

    --
    ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
    Starting Score:    1  point
    Karma-Bonus Modifier   +1  

    Total Score:   2  
  • (Score: 2) by Reziac on Sunday July 19 2015, @04:55PM

    by Reziac (2489) on Sunday July 19 2015, @04:55PM (#211117) Homepage

    I usually pick up a used Enermax PSU off eBay for $25 or so, they run forever and you can load 'em right up to the edge of their stated capacity without a hitch. I have a boxful of salvaged lesser PSUs, useful for this and that, like you say good for when you're not really using all that much juice (and maybe don't care about longevity -- motherboards don't last so well when the PSU is crap). But my everyday systems get an Enermax. My UPS shows actual draw and it's interesting that at idle, the others wander all over the place, but an Enermax will be totally steady. And an Enermax will be made =by= Enermax, not a rebadged gods-know-what.

    BTW if you ever need an AT PSU, TOPower still makes one (you can order it direct) and for $65 it's still top-notch. The one in my old everyday box got lightning-fried (thru two layers of surge protection and a UPS) and since I still needed that box ... can't tell the new PSU from the 300W I'd bought 15 years before (and ran 24/7 at max load all those years, replaced the fan a few times), except being a 400W it has more connectors. It weighs just under 5 pounds, honkin' big capacitors and heatsink.

    That's unfortunate about the POST testers now. I've gotten assloads of use out of my old one. Works great for pinpointing an invisible problem, or when the system doesn't get as far as ANY beeps or looks dead to the naked eye but really isn't. And when I've bothered to dig deeper, the POST tester has always proved right. Yep, usually nothing we can fix, but handy as hell when the problem isn't something obvious like bad caps. Or when it's bad caps that don't *look* bad. 15 seconds to pinpoint the problem, instead of half an hour connecting and disconnecting shit to find what's being cranky, what's not to like? Also great when you're trying to get an RMA, they believe you right off instead of wasting your time making you check shit you know ain't the problem but is in the tech's script. Well, being it's about $10 for fresh one, not like it's a major investment, and I'll bet my old codebook still matches current codes. Doesn't need to be any more specific than, say, "06 = IDE channel" or whatever. (I used the old one so much I had all the codes memorized, but have since forgotten 'em.)

    --
    And there is no Alkibiades to come back and save us from ourselves.
    • (Score: 2) by Hairyfeet on Monday July 20 2015, @09:48AM

      by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Monday July 20 2015, @09:48AM (#211353) Journal

      Sadly I rarely get the luxury of using a "used PSU" because the customer doesn't care if its a $120 PSU you are getting them for $35, all they care about is the "used" part. I've had a customer turn down a $100+ Coolermaster gamers PSU used for $30 to to take a shitty $40 550w HongFong fuckery because the crappy was NIB. That is why I grab the Apevia, you can get their "850W" (translation 550W max) for $40 and lets me show the customer its NIB. Most of those types are using some Athlon X3 or C2Q box that ain't gonna pull more than 300W max for its entire lifetime so if they want only NIB and won't shell out for decent parts? Meh here ya go, take some pretty shit with the blue LEDs and crap LOL.

      And yeah it sucks but I think what happened is shops went from testing gear to just tossing it and when combined with the race to the bottom mentality most of the good tester gear either quit being made or went to the ultra high dollar server market. the last shop I did some work on the side for had a fucking bad ass board tester, this thing would not only test the unit but plugged into the 24 pin and CPU pin slots and would power the board as it tested it (to give it clean power) and had a tester for both PCI and PCIe. Sounds nice huh? It was....cost nearly $900! for the thing too. Considering I can get new boards for most sockets under $50 spending that much on a tester when its gonna tell me that 90% of the boards are just junk? Yeah just not worth it. the one I linked to will read off the standard BIOS error codes but with everything going UEFI I don't know how long it'll be useful to ya, but its only $7 so if it only helps on a couple boards a year you'll get your moneys worth.

      On new boards I stick with Amazon and Tiger as I've never had a bit of trouble getting an RMA without jumping through any hoops. This is why I try not to buy anything from Newegg I can't afford to just chunk in the trash, they can be the worst when it comes to getting an RMA. That said I fucking looooove the case I got for Xmas [newegg.com] as its a fucking beast, toolless everything, sound dampening on the drive cages, and big fat fans that keep everything icy without having to spin fast and whine. I got an FX8320E (love them E CPUs ya know) with 16GB of RAM on an Asus board, R9 280 and 6TB of HDD with an SSD boot stuff in that thing, quiet as a churchmouse. Of course Rosewill is a Newegg brand so I had no choice but when it comes to cases for myself I like big solid tanks with tons of elbow room and the reviewers were right, its solid.

      As far as AT goes? I'm lucky in that I get tons of PCs from the building super (who also works for the city so can grab anything he wants from the recycling center) so I'm usually just swimming in parts. The download box I use at the shop? Fricking nice HP Pro 3000 with 4GB of DDR3, C2Q and 500GB HDD that was still in the plastic some rich bitch was tossing, the reason? Hubby got her a Macbook. Ran HDDTune and checked the drive time and the thing had run for less than 30 minutes since it has been bought,never underestimate the ability of rich folks to waste money. Oh well makes a damned good office box and like a lot of parts at the shop cost me $0.00 LOL, gotta love it. Need to contact my buddy across the state that takes my old junkers, starting to get a few too many P4s in the pile and I won't touch those pigs, even the Pentium Ds I instantly drop on the scrap pile as netburst was just too fucking slow and power sucking, I HATED those damned things. He has no problem selling cheapo "granny boxes" so I let him haul off my junk just to keep it from getting dumped in the garbage.

      Anyway that one tester is just $7, its worth a shot even if they aren't as good as they used to be. Just be sure to get that PSU tester as that thing works perfectly and getting more than a "pass/fail" really helps you decide whether it needs chunking or just needs put in an office box, well worth the $12 IMHO.

      --
      ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
      • (Score: 2) by Reziac on Monday July 20 2015, @01:40PM

        by Reziac (2489) on Monday July 20 2015, @01:40PM (#211408) Homepage

        I don't imagine you'll have a customer who needs AT unless they're running some industrial equipment that only speaks to stone-age hardware. AT is otherwise so long gone that after the P233s junked by the local school system we never got any at the user grope. My old Tyan boards swing both ways but are happier with AT than ATX, and they're special so what they want, they get. :)

        Yeah I figure for under ten bucks, if I get a couple good uses out of the POST tester then it's paid for itself. Ordered that and the PSU tester, added some cable adapters I needed anyway so I'd get free shipping, then discovered I had a $25 gift card from gods know where plus a $10 coupon I don't recognise either, and my whole order was just about paid for. :)

        Me, I'll take the good used part over the crappy new part any day, and as someone once said of cars they all run on used parts. Just picked up another lightly-used Enermax 600W PSU for cheap (the last "used" one I bought off eBay was shiny-new, this one isn't much worse), could use one or two more of these when they come along at the right price but no big rush.

        I have cases coming out my ears, when I moved I only kept the ones with lots of drive bays and easy to work in, and still have more than I need. I'm not a big fan of what I've seen in toolless, don't have the control over exactly where things line up that I like to have. But speaking of fans that's another thing I need to order a few of... down to no extra case fans and need to replace this Sunon that's where I can hear it; the durn Sunon fans NEVER fail (great for putting in somewhere that's a bitch to get to), but they sure are noisy. What's reliable, cheap, and quiet nowadays?

        Good to know about the RMAs. Tiger used to be the worst in the business, but from what I've been hearing they seem to have turned things around.

        In my museum I have a Mac G4 someone paid $4000 for and far as I can tell, never used AT ALL... yeah, makes you wonder. (Except that given the MacOS9 interface, I really don't wonder. They probably were utterly baffled and that was the end of trying to use the damn thing.)

        --
        And there is no Alkibiades to come back and save us from ourselves.
      • (Score: 2) by Reziac on Monday July 20 2015, @04:04PM

        by Reziac (2489) on Monday July 20 2015, @04:04PM (#211469) Homepage

        Found my notes on the CPU on the M2N68-AM Plus.... AMD 4850e 2.5GHz. 45W so not bad there!

        Someone's comparative benchmarks:
        http://www.cpu-world.com/CPUs/K8/AMD-Athlon%20X2%204850e%20-%20ADH4850IAA5DO%20%28ADH4850DOBOX%29.html [cpu-world.com]

        Doesn't look like I'd gain much by replacing it, and everything else drinks more juice.

        --
        And there is no Alkibiades to come back and save us from ourselves.
        • (Score: 2) by Hairyfeet on Sunday July 26 2015, @08:50PM

          by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Sunday July 26 2015, @08:50PM (#213970) Journal

          If it were me I'd look at an AMD Phenom X4 9150e [ebay.com] As that would only raise the power by 20w (and that is max, I've found daily usage on one of these to be in the 20w-25w range according to kill-a-watt) and if you keep an eye on ebay you can score them for as low as $10-$15. The 4850e is an okay chip if all you are doing is web surfing but if you actually want to do any heavy lifting, DVD conversion, transcodes, A/V work, then you will quickly run out of oomph with the dual cores.

          And I'm glad the POST tester worked out, between that and the PSU tester you should be good to go.

          --
          ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
          • (Score: 2) by Reziac on Sunday July 26 2015, @09:26PM

            by Reziac (2489) on Sunday July 26 2015, @09:26PM (#213982) Homepage

            Well, the AMD box isn't intended for everyday regardless... the two Intel quad-cores are 30-40% faster than the best that AMD socket supports, per CPU-world's benchmarks and per checking 'em with Everest. But I'll keep an eye out for a freebie that it can use. :)

            I thought the quad would eat more, but my UPS shows actual draw, and with 3 IDE HDs, two opticals, floppy, card reader, 6GB RAM, the 1GB Radeon, and two extra fans -- 89W at idle, 165W running the CPU full-bore.

            For comparison, idling:
            AMD3200+ (one optical, one FDD, one IDE HDD, one SSD, 4GB RAM): 96W
            AMD4850e (2 opticals, one FDD, one IDE HDD, card reader, 2GB RAM): 64W
            Intel Q8300 (2GB RAM, nothing else attached or running except the CPU fan): 56W

            (Need to find that Q8300 a better mobo... Dell's Foxconn mobo's onboard video maxes out at 8mb!! probably any mobo picked from the trash would be an upgrade.)

            That Radeon vidcard sure has stupid fan design, the fan just hangs from the bearing race (blade unit isn't actually mounted on anything!) so it rattles something awful. Disconnected it and tap-screwed an old 486 CPU fan to the heatsink, and it's no worse off for heat and a hella lot quieter.

            --
            And there is no Alkibiades to come back and save us from ourselves.
            • (Score: 2) by Hairyfeet on Monday July 27 2015, @12:12AM

              by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Monday July 27 2015, @12:12AM (#214026) Journal

              That "30%-40%" is not surprisingly almost exactly what the Intel Cripple Compiler slows down the AMD chips by thanks to using X87 math ONLY instead of using the SSEs and other new math paths. You fire up programs that aren't cripple compiled? You find that until you get to the Intel chips that cost over $600 a pop the average difference is between 3%-6%, on the chips you have its even closer, we're talking 2%-5%. For an example of how fucking badly Intel has been rigging things just check out the Linux benches [phoronix.com] or the ones from Tek Syndicate [youtube.com].

              The moral of the story? Unless you are ONLY gonna run benches and programs run through ICC (which luckily for the world the majority of programs are not compiled with the crippler) then you are gonna find that 30%-40% just does not exist and the difference between that C2Q you have and that Phenom X4 I linked to are practically nonexistent. Last I checked handbrake is compiled with GCC so if you wanna run a little informal test yourself encode a video with HB and time it, either cutting the time in half on the 4850e due to it being 2 cores short or even easier just turn off half the C2Q cores by setting the affinity on HB before running it.

              And if the Q8300 board has a PCIe the answer is obvious, just watch Tiger for one of those $10 GPU deals. if its an AGP then ebay is your best bet, the Radeon 2400 can be had for around $8 and that is great for just running video, or a Geforce 6200 will likewise be better than onboard while being dirt cheap.

              --
              ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
              • (Score: 2) by Reziac on Monday July 27 2015, @01:21AM

                by Reziac (2489) on Monday July 27 2015, @01:21AM (#214039) Homepage

                Actually I was going by CPU-world's benchmarks, which use a broad range of tests. For hands-on I use Everest, but it's always pretty much reflected realworld performance (I can detect differences as small as 3% myself, just eyeballing -- I have a ridiculously accurate internal clock). And it's a good match for what I've observed among the "new" frankenputers, to where I'd say CPU-world's marks are dead on.

                That Foxconn board has one PCI and one PCIe, probably anything that comes along would be better'n its onboard video (tho I'll give it this, it has good color) but a better motherboard entirely would be better yet! Or could be it'll get some grunt job like ripping discs or serving video over the network, where no one cares if it can even make a picture. That's what I did with the last OMG-this-is-junk-but-it's-fast-junk mobo, made it useful til it died.

                --
                And there is no Alkibiades to come back and save us from ourselves.
                • (Score: 2) by Hairyfeet on Monday July 27 2015, @06:57AM

                  by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Monday July 27 2015, @06:57AM (#214151) Journal

                  Well I know Everest and Sandra are paid by Intel to use the crippler so are frankly worthless for testing, don't know what compiler CPU-World uses but if they do not list the compiler I would consider it suspect. If you haven't looked up the tests on how bad the crippler rigs tests you really should, one researcher managed to get a low end Via to "magically" score 35% higher on Sandra benchmarks by simply changing the CPU-ID from centaur hauls to genuine intel. this is why I refuse to sell new Intels in my shop, because if you win fair and square? You deserve the rewards and accolades for a job well done. If you OTOH rig the outcome? Then they should be treated like plague blankets, as rigging benefits nobody but the one doing the rigging. Just look at the price of the i7s trading blows with the FX8350 I linked to, if the tests weren't rigged in most places Intel would not be able to keep their prices 300% higher on those chips.

                  As for the GPU? Keep an eye on ebay and newegg for the Geforce 210/310 units. They can usually be had for $10-$20, most are silent, and with a GB of DDR 3 they are really great for cheap boxes that lack decent video. Of course if you want to spend nothing just slap on a copy of Win 7 Tiny and use her up. the only thing I didn't care for design wise with the C2Q Q series was the lack of multiple speedstep levels, the box I have at the shop for general purpose is a Q6600 and with an HP mobo its only got 2 speed states, 1600 and 2400 mhz which translates to 119f and 140f when running full bore doing transcodes and that is despite having an aftermarket heatpipe installed. Say what you want about the AMD E series but the temps were a good 10f-15f cooler when I had one of those...man I really shouldn't have sold it LOL.

                  --
                  ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
                  • (Score: 2) by Reziac on Monday July 27 2015, @01:14PM

                    by Reziac (2489) on Monday July 27 2015, @01:14PM (#214287) Homepage

                    Methinks the CPU ID compiler flag is being mistaking for deliberate crippling. And it just ain't so. It compiles to reflect a given CPU type's instruction set. AMDs lack some of the Intel instruction set, and have some of their own. You can set the flag however you like. Surely there's some opensource test suite a person could experiment on using the GCC (opensource) compiler and prove that for themselves. Hell, just use GCC and some big ugly codebase like Mozilla as the test case, that ought to take long enough to test basics with, no need for a benchmark app.

                    As to benchmark twiddling, AMD has hardly been a saint there. They've lied about mathco performance since day one, that being where they generally fall down compared to Intel (and still do). And somehow they always do fantastically better than Intel on gaming-related marks even if they're way behind everywhere else (noticed this on CPU-world too). This leads me to strongly suspect AMDs offload some of their work to the GPU on the sly, being one of the tricks of gaming performance coding, but the benchmark can't see it happen cuz it still passes through the CPU.

                    And comparing errata (once AMD finally started making theirs available, which they didn't for years) is always an educational experience. (just search for the CPU name plus 'errata'. First example I came to: http://support.amd.com/TechDocs/48671.pdf [amd.com] ) AMD's errata list is generally 3x as long as Intel's, and "no fix planned" seems to be their main response to their own bugs. The erratum that cured me of buying AMD caused 32bit OSs to run in 16bit mode, talk about a performance hit!! mind you this was 1997 so that was pretty durn late to have that be excusable as an early bug, and they never did officially admit to it (I had insider info about it).

                    This here Q6600 is running at 100 to 120 degrees, depending which sensor you look at (apparently between CPU and mobo, it has four). The mobo is an Intel DQ965GF, Intel stock CPU HSF, PSU has two fans and case has an extra intake fan, plus the unused upper front drive bays are open (which improves the average box by about 10 degrees all by itself). Most of the heat comes from the Radiator, er, Radeon, which has several sensors and runs at 120 to 180 degrees depending which one you look at (prolly was worse with the cover on and the stock fan). The two AMD CPUs are in the same temp range also with stock HSFs, no big advantage there either way.

                    That Dell mobo doesn't even have IDE or FDD ports, but is fairly smart about booting from USB, so at least it's not devoid of virtue. But what's with having four SATA, 0, 1, 4, 5 but ports 2 and 3 are missing? can Foxconn not count to six without missing a few??

                    --
                    And there is no Alkibiades to come back and save us from ourselves.
                    • (Score: 2) by Hairyfeet on Monday July 27 2015, @06:16PM

                      by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Monday July 27 2015, @06:16PM (#214447) Journal

                      Nope sorry, that was already proven false, how? The very first crippler targeted the Pentium III and NOT the AMD! That is right, before they turned the crippler on AMD they used it on their own chips first! Why would they do that? Because the Pentium III was scoring nearly 40% higher on most benches while costing less than the Pentium IV, they released the crippler, paid the major benchmarking companies to use it and wadda ya know, now with the exact same chips suddenly the Pentium IV is scoring nearly 40% above the chip that was stomping it! Amazing huh?

                      If you do a little research into the crippler you'll find its been documented by researchers going back to 2000. It was evidence of the crippler that got Intel to shell out 1.25b to AMD (which I still think was a travesty, they should have gotten antitrust like MSFT did) and the P III is not only a smoking gun, but is also evidence of just how badly the crippler cripples code. You see there hasn't been a single chip released by either company that doesn't have SSE 1,2,3 in over a decade which are standardized and identical specs with both companies, right? Well guess what the crippler uses when it detects AMD by CPUID? NOT SSE 3, nor 2, nor even 1, it uses X87 math paths which was depreciated nearly 3 years before the crippler even came out!

                      So they aren't just rigging friend, they are rigging the living fuck out of the tests as the X87 code path was written waaaaay back when CPUs used an X87 co-processor. This is literally as low as they can possibly go and still put out code that will run, and between that and the fact that the crippler will not even check for the standard SSE extensions? yeah there is no doubt, it was made for the express purpose of rigging, no different than "quack.exe" back in the day.

                      --
                      ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
                      • (Score: 2) by Reziac on Monday July 27 2015, @07:28PM

                        by Reziac (2489) on Monday July 27 2015, @07:28PM (#214483) Homepage

                        See, the problem I have with that is there are assloads of compilers, some have CPU-specific optimization and some don't. And you can't count on everyone using the same optimization flags just on Intel's say-so, far from it. (Imagine linux fanboys, most of 'em being AMD fans to boot, doing any such thing... whoops, which OS wouldn't run at all on those buggy AMD chips, back-when? yeah, linux.) If you've got links I'll be happy to read 'em, but I'll also take into consideration whether the source knows what the hell they're talking about, cuz it sounds to me like they're misinterpreting what compiler optimization routines DO, never mind having any clue how many compilers there are to choose from. You can't optimize using a path your host CPU doesn't know how to do, or does so buggy that it's best avoided. Anyway, like I said I'd be glad to read whatever you can point me at, won't hurt my feelings either way, but I'll chase down my old programmer buddy over at OpenWatcom and get his opinion before I take any of it too seriously.

                        --
                        And there is no Alkibiades to come back and save us from ourselves.
                        • (Score: 2) by Hairyfeet on Tuesday July 28 2015, @12:55AM

                          by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Tuesday July 28 2015, @12:55AM (#214634) Journal

                          But SSE 1-3 was written by Intel so to have that standard it HAS to be identical. Now if they have to have SSE as documented by Intel to say its an SSE chip...why would they use X87, which is literally the slowest code path they can choose and still make functional code? The answer why is obvious, its to rig benchmarks. BTW you might like to know Intel admitted to rigging benchmarks [youtube.com] WRT to Pentium 4 as part of a class action settlement. If the company comes right out and says "yep we did it" I really don't know what more proof you need, they flat out admitted they rigged the benches to make P4 beat Athlon when they knew P4 was a dog.

                          Oh and you might want to check out the various "remove compiler bug" patches out there, researchers have simply patched out the CPUID check on several programs and found the programs run perfectly with the standard SSE codepath with ZERO penalty or crashes, whether its on AMD or Intel. So again why do it, if its only purpose is to slow down the competitor OTHER than you want to make sure any contest is rigged? You can take any FOSS program, compile with ICC and GCC and even though GCC is frankly not nearly had the money sunk into it that ICC has you'll find that the scores? Damned near identical between Intel with ICC and Intel with GCC, we are talking maybe 4%, low enough to be margin for error. So ICC isn't making Intel chips run any faster, all its doing is making AMD chips run slower which is why I hope the current EU investigation nails Intel to the wall, because everything indicates they have been rigging benches since the first P4 release.

                          --
                          ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
                          • (Score: 2) by Reziac on Tuesday July 28 2015, @02:03AM

                            by Reziac (2489) on Tuesday July 28 2015, @02:03AM (#214652) Homepage

                            Well, that guy in the video is pretty durn vague, not what I'd call definitive. I don't come up with anything for "remove compiler bug" (someone consolidating a GPU driver was the closest match, otherwise nothing). And I've had a whole range of P3/P4 to compare and I'm not seein' what they're bitching about. There's a bigger performance jump between 2GHz and 3GHz than it looks like there oughta be, but you kinda expect a tech to improve as it matures, yes? first few iterations of anything sucks. Sometimes there's no fixing the suck and you back up and start over. But you never admit that in marketing, cuz that's kiss-of-death even if it was the best thing to do.

                            Back-when I did see AMDs bench well but fall down badly in actual everyday use, the last one I was stuck using myself being (now I feel old) a K6-2 450MHz, and it wasn't even close to the supposedly-equivalent P2/P3, in fact my old P233 ran rings around it, it was embarrassing... and as I recall there was a lot of question about how they got those K6-2 benchmarks to look so good.

                            Anyway, I don't think anyone is simon-pure here, they're all out for their own advantage, but it's got a ways to go to convince me there's that kind of wrongdoing, and convincing me the performance I can see with my own damn eyes is wrong? That's a real tough sell. I've got my own questions on the current AMDs benchmarks, cuz like I said it looks to me like they're cheating at certain points themselves, there's no way they're suddenly that much faster on gaming-specific functions but simultaneously that much slower at heavy math, but offloading to the GPU and counting it as CPU performance? could be.

                            Tho truth is I don't care that much who's truly faster than whom, cuz nowadays I'm about as likely to buy a new CPU off the shelf as I am to win the lottery, so whatever comes along as salvage and runs stable is fine with me! Too bad you're not close enough to pillage your scrap heap, prolly lots of stuff I'd cheerfully haul off and put to good use. :)

                            As the W.D. drive capacity class action suit should have amply demonstrated, the courts are not a quality place to determine blame or truth when it comes to tech stuff, especially when juries are cherrypicked for ignorance and swayability and tech ain't the judge's field either. And what comes out of the EU courts is sometimes a whole different class of crazy.

                            --
                            And there is no Alkibiades to come back and save us from ourselves.
                            • (Score: 2) by Hairyfeet on Tuesday July 28 2015, @07:35AM

                              by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Tuesday July 28 2015, @07:35AM (#214742) Journal

                              It looks like my post didn't go through, so sorry if it ends up being a dupe but here is the article [pcworld.com] and you can even click on the link and get your $15 as part of the settlement. I already filed for mine BTW.

                              They have been caught multiple times rigging, one Toshiba exec admitted the kickbacks they were getting not to sell AMD chips was "like cocaine" and Michael Dell himself admitted that there was quarters during the great PC price wars where the only Dell profits were Intel kickbacks and that was in court during the discovery phase of the AMD lawsuit which they shelled out 1.25 BILLION to quash...what more proof do ya need?

                              I myself was a diehard Intel+Nvidia guy...UNTIL it came out how badly the two corps was rigging, now not a single chip by either corp will be bought by my shop! Did AMD have some bad runs in the past? Sure, so did Intel, anybody that had an Abit or MSI will tell ya that, and I have to wonder how many of AMD's problems were because Intel had rigged the market so badly that no board OEMs were willing to spend top dollar developing AMD boards knowing no OEM sales would come from it? I know that in their SCC filing they stated that one of the big reasons for buying ATI was to have control of the quality of the chipsets, and if you were dealing with K6-2s you know how bad those were. Remember those Via chipsets? Shudder.

                              But I can tell ya right now I'll happily take the Pepsi challenge against any Intel chip 2 to 3 times the price of my FX8320E and have zero doubt I'll win, just sitting here watching Judas Priest in concert while typing this AND using Handbrake to transcode a bunch of old vids for my mom convinces me of that. Know what the highest wattage I've gotten my board to is? 67w and that is with all 8 cores slammed to the firewall, on regular usage I'm only pulling between 8w and 16w. Oh video just got done, current pull for 8 tabs with a 1080P YouTube vid going? 23w at 1400Mhz.

                              If you wanna support a company that admits they are rigging the game? That is your choice, but I believe in a free market and a fair game, so Intel is brought to task (hopefully by the Eu antitrust investigation) and is forced to stop stacking the deck? No new Intel chips will cross my door. But if you decide you wanna max out that 4850E board on the cheap? Keep an eye out for the Phenom I quad Es on ebay, or hell grab one of the first gen Phenoms, I've gotten them for as low as $8 thanks to the TLB bug and I can tell ya from selling dozens of those that you got better odds hitting the powerball than hitting that bug. Heck when that bug was announced and the Phenom IIs hit I was buying those chips by the fistload and making everything from office boxes to gaming systems out of 'em, never once did I hit the bug.

                              --
                              ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
                              • (Score: 2) by Reziac on Tuesday July 28 2015, @03:14PM

                                by Reziac (2489) on Tuesday July 28 2015, @03:14PM (#214876) Homepage

                                I read the PDF. Looks damning on the surface, but it's all undocumented allegations and quotes out of context. That it's a class action suit and not an actual investigation causes me to eye its credibility and interpretation-of-events well-salted and from a safe distance.

                                It's important to remember that class action suits are not about exposing wrongdoing or making things right. They are entirely about making lawyers rich (one plaintiff gets rich as a byproduct, everyone else gets a token settlement and ultimately higher prices -- where d'you think the money comes from to pay for all this??) Remember that the lawyer got 30-50% of the core settlement, in cash. As a rule the lawyer picks the target (here being Intel), then basically hires someone to be the plaintiff (get 'em to claim they were wronged and they both make out like bandits... funny how sometimes a given lawyer and plaintiff have an ongoing relationship). Class action suits are so inherently crooked as to invalidate whatever they're supposedly meant to redress. Remember there is no one investigating whatever evidence is presented, other than the defense's lawyers. There's absolutely no independent corroboration of anything. There's no way of knowing if maybe AMD and the lawyer had an understanding on the side, either. You sue them, we look good in the marketplace.

                                Intel was the target because they had the deep pockets. If AMD had the deeper pockets, they'd have been the target instead (doesn't take much to give a class-action suit an opening that will hold up in court, because it's not about proof as we'd think of it, only about convincing the court to make someone pay, not so hard in today's legal climate). Consider that this is about events going on 15 years ago, and if that's really relevant to anything today. One reason you do class-action over old events is that by now a lot of the evidence that could nix your suit is gone to the bit bucket -- makes it hard to disprove even if what you've got is totally out of context. Sometimes it comes out years later that the evidence presented was bogus, too (eg. Erin Brockovich case).

                                And once there's a class action suit, the company being sued *cannot* admit wrongdoing, even if they might otherwise, cuz that would be legal suicide.

                                I suspect if you had access to the same level of documents out of AMD, you'd be twice as shocked (especially considering how much massaging of their gaming benchmarks was going on back in the K6/K7 era, tho looks like more of same today, to me). I can tell you for a fact they hid fatal bugs like the won't-do-32bit-OS bug I mentioned before. Ever wonder why Windows insisted on running in 16bit mode on some AMD boards? That was it. I wouldn't know about it either except a 13 year old friend had nothing better to do than harass AMD engineers til he got an explanation (and a replacement chip) ... under the table. "You didn't hear it from us." Back then, AMD didn't publish errata at all.

                                That there wasn't as much effort put into AMD boards? Not so from what I saw -- when you get high-end companies like Tyan supporting AMD in their server-class boards, well, that argument kinda breaks down. Cheaper tends to attract cheaper, and AMD was pursuing the lower end, so we got all those AMD+VIA pieces of shit to satisfy the cheapskate end of the market, but that's the same market force that puts cheap all in the same box together everywhere, not just tech. Go to Walmart and look at cheap tools, you won't find any German carbide tips in the made-in-China set.

                                Dunno if it's the bug you're talking about but about 5-6 years ago I was at a friend's shop and he was having hell's own time with an AMD64, not sure what model but whatever was their top-of-the-line at the time, on a primo Tyan server board ... put a high-end nVidia card on it and use Windows magnifier, and the thing would lock up every time. Put an AMD32 on the board, or anything but an nVidia, and the problem went away.

                                --
                                And there is no Alkibiades to come back and save us from ourselves.
                              • (Score: 2) by Reziac on Wednesday July 29 2015, @01:59AM

                                by Reziac (2489) on Wednesday July 29 2015, @01:59AM (#215186) Homepage

                                Hey! got the power tester. Works a treat. Applied it to my whole boxful, weeded out a couple, nodded a couple times on units I'd thought were iffy (oho, that one I'd marked "iffy" is spiking, that's what's wrong with it.. and that one I'd marked "weak" can barely do 10v on the 12v rail). Became even more of an Enermax bigot... runs closest to spec across the board, and it's more stable under load than with nothing on it (was handy to check with a couple HDs plugged in, so I did). Now to spend an hour cannibalizing fans out of the rejects. Well, I needed a few fans...

                                Anyway thanks for pointing me at it -- dandy little tool. You're a pal. Have a beer!

                                --
                                And there is no Alkibiades to come back and save us from ourselves.
                                • (Score: 2) by Hairyfeet on Wednesday July 29 2015, @11:32AM

                                  by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Wednesday July 29 2015, @11:32AM (#215375) Journal

                                  Sorry I can't touch alcohol, gives me a headache :-(

                                  The bug I was talking about is the "TLB Bug" which like the old Pentium II floating point bug you have better chance of hitting the lotto, only in the AMD case it was even harder to hit. Basically certain kinds of math (the kind a normal person will never use BTW, as I was never able to hit it even doing heavy transcodes) will cause the TLB buffer to dump which slows down cache look ups...big fricking whoop, especially when that bug meant you could buy quads for under $40 then and less than $12 now!

                                  And its a hell of a slog but if you want all the proof you'll ever need that Intel rigs? Look up the court transcripts for the AMD VS Intel case. During the discovery phase they interviewed a LOT of high level guys from the big OEMs and they all gave the same story, Intel threw them shitloads of money to NOT carry anything but low tier AMD chips, likewise they paid the big benchmarking groups (which they still do to this day BTW, all the DoJ required is they stick a "warning" in ICC that basically says it sucks for anything but Intel chips) piles of cash in "advertising incentives" to only use ICC with their suites. this is why the Linux sites had wildly differing scores then and now, they use GCC as a matter of course so no ICC rigging.

                                  And the bug your friend ran into was NOT with AMD, it was the fact that a lot of the tools in XP were using crap code from the Win9X days and 16bit code didn't work on 64bit CPUs. This is why I pushed a lot of clients to XP X64, it was 100% legacy free thanks to being Server 2K3 X64 with an XP skin. Now of course we don't have to worry about that since MSFT ditched the legacy crap with Vista, we only have to worry about the frankenstein mess that is the Win 10 settings/control panel monster LOL.

                                  Anyway glad the tool worked out for ya, you'd be surprised at how many weird errors can be traced back to crap power. The nice thing about that model is unlike your old "pass/fail" you'll find a LOT of PSUs that it would give a "pass" to IRL were right on the line or suffering from serious drops. Have fun ripping apart all those PSUs LOL!

                                  --
                                  ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
                                  • (Score: 2) by Reziac on Wednesday July 29 2015, @12:20PM

                                    by Reziac (2489) on Wednesday July 29 2015, @12:20PM (#215393) Homepage

                                    The client needed 64bit. The common factor for any lockup was the AMD64 chip. Couldn't MAKE the system fail without it. Could lock it up 100% reliably with it. With the AMD64 on there, pretty much any messing with the video would lock up but the magnifier was a quick and easy test that didn't fuck with anything (like configuration stuff).

                                    I put XP64 on the "new" box, cuz, 6GB RAM that fell on my head. Some things are definitely better'n XP, tho I find myself missing small fixes from SP3 here and there, and I'm gonna have to find some solution less awkward than DOSBox for a few old apps I still need.

                                    Well, if I can't buy ya a beer, how about an ice tea or a lemonade? :)

                                    --
                                    And there is no Alkibiades to come back and save us from ourselves.
                                    • (Score: 2) by Hairyfeet on Wednesday July 29 2015, @07:30PM

                                      by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Wednesday July 29 2015, @07:30PM (#215587) Journal

                                      Then I bet it was a bad chip, look at the errata on either Intel or AMD sometime, a lot of little "quirks" means that bad chips DO come off the line sometimes. I once had a customer whose computer was constantly corrupting, turned out his particular P4 was bad. Basically any math more complex than your basic maths would cause all kinds of weird returns you could put in a complex equation in the thing and it would give you a different answer every.single.time. Weirdest shit. Ran into a similar problem with an AM2 Sempron owned by my former landlady, everything else worked beautifully but that Sempron would corrupt the OS over time. tossed it for an Athlon dual? She is still using it to this day.

                                      Funny that you should mention tea as that is what I'm having right now, we southerners gotta have our sweet tea ya know ;-)

                                      And have ya tried virtualbox? Its been awhile since I ran Win9X on it but VB worked perfectly with 98SE last time I needed it and Win98 was so light on resources pretty much any box in the last decade can run multiple Win98 instances without breaking a sweat. Of course if you are always getting old boxes you might want to just slap together an Win9X box, you should already have a KVM (and if you haven't got one, why? They are fricking GREAT!) and pretty much any old thing will run Win98 which gives you native DOS as well as all those Win9x programs that don't run anymore. For years I kept a SFF Compaq 733MHz with a TNT 2 for a Win9x box, it was small enough it could be propped on its side behind a desk and was ready for whenever I felt like a little Mechwarrior 3. You can pick up a 4 port PS2 KVM for a little of nothing and its certainly easier than dealing with DOSBox.

                                      --
                                      ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
                                      • (Score: 2) by Reziac on Wednesday July 29 2015, @08:55PM

                                        by Reziac (2489) on Wednesday July 29 2015, @08:55PM (#215606) Homepage

                                        Yeah, it's probably time to try VirtualBox. With 6GB RAM in the XP64 box I have no excuse not to, haha. Actually I'd like to clone my entire old Win98 setup, DOS utils and all, and have that available in a VM. I still have its hardware (for that matter I still have my Win95 box intact, among others) but seriously, how many machines do I need running at once? :)

                                        DOSbox makes me crazy, what with the mount/unmount crap. I guess there's an easier way to make it load stuff automagically, but I'll have to look it up again.

                                        I've had a Belkin 4-holer KVM since forever, must be 25 years old, it has both PS/2 and AT keyboard ports! but it won't speak to these newfangled flatscreen monitors at all, not even the ones that still do regular analog VGA.

                                        Have had a couple cheapie KVMs since but one issue or another with 'em, like can get video and mouse to work but not keyboard. And seems they are actually USB devices despite being connected via PS/2 and vidports, and draw a LOT of power, didn't realise that til one day Windows bitched that "a USB device is drawing power beyond its specs" and shut down the KVM! My old KVM has a wall wart, but these newer cheapies don't.

                                        Need to get one that will speak to a USB-wireless mouse, and a 4-holer, but don't want to spend a fortune on it. ConnectPro are supposed to work best but I really don't want to spend that much. Had in mind more like this price range but no idea if the unit is any good.
                                        http://www.ebay.com/itm/390773804184 [ebay.com]

                                        With regular USB connectors, I've found you can power ANY device that's on a hub with a wall wart and everything on the hub will draw from it, so that might be a workaround if that drawing-too-much-juice is a chronic issue. (Figured that out when my unpowered hub was suddenly powering everything hooked to it even when it wasn't connected to the PC... cuz the one USB HD has its own wall wart and the hub was evidently drawing off that.)

                                        Always possible that AMD64 was a bad chip, but when they're still an $800 chip you don't exactly keep spares in inventory, so that theory was kinda hard to test on the spot. But as it happens he went looking online and turns out he wasn't the only one affected; it was indeed a known bug. Much swearing ensued. I don't know what he finally ended up doing about it.

                                        I've read some of the errata sheets, and sometimes you wonder how the hell something so obvious sneaked past the engineers!!

                                        And now I think I'll have a cup of min tea in your honor :)

                                        --
                                        And there is no Alkibiades to come back and save us from ourselves.
                                        • (Score: 2) by Hairyfeet on Thursday July 30 2015, @05:36AM

                                          by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Thursday July 30 2015, @05:36AM (#215753) Journal

                                          Well if you find a good 4 port USB KVM, will ya let me know? Everyone I've looked at under $100 has serious "gotchas" like not picking up the EDID so your screen is boned every time you switch or Windows not reading on boot or complaining something is unplugged when you switch. I have a couple year old PS2 KVM, Trendnet I think (not crawling back there to look LOL) that works perfectly with my 1080P monitor, so much so I'm seriously thinking of plunking down $30 for a USED gaming mouse so I can have more than 3 buttons supported by PS2. you would not believe how fricking hard it is to find a PS2 mouse that has more than the standard 3 buttons, just nuts.

                                          BTW you will want to AVOID that unit unless you have space on your deck for it! I had one just like it, ended up giving it away as they do NOT have hotkey support! I don't know about you but I have enough shit on my desk so if I can't go "scrolllock scrollock (numberkey)" its just not worth a piss to me. For some damned reason the USB KVMs are kinda notorious for being buggy, why its so damned hard for them to simply emulate a keyboard and mouse is beyond me but I've looked at a couple dozen so far and its all the same, hassles and BS.

                                          And yeah not having a testing unit does suck, I would have probably gone to my nearest repair shop and seen if they had one they'd let me use. You'd be surprised how easy it is to get help from most shop guys if you just bullshit with us for a bit, we are usually bored so somebody that can talk the lingo? Will usually get extra consideration.

                                          But if you have any luck finding a decent USB KVM let me know, been running KVMs for 20 years and just can't go back to running singles, ya know?

                                          --
                                          ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
                                          • (Score: 2) by Reziac on Thursday July 30 2015, @04:30PM

                                            by Reziac (2489) on Thursday July 30 2015, @04:30PM (#215924) Homepage

                                            You gave away one like the cheapie KVM I linked to, or the ConnectPro? I can live without a hotkey -- if there's gotta be just one or the other I prefer having the switch, cuz that still works if for some reason it's lost track of a keyboard (my old KVM has both). It's gonna go behind my monitor regardless, not so far to reach. Might help the small clutter stop accumulating there, haha. Cables reproducing like snakes instead, that's the ticket. What lies behind every great computer? A mess of wires! :)

                                            I thought it might be that the antique KVM doesn't know over 1024x768, but that's the default on my workbench flatscreen, and it doesn't like that one any better. It's just not getting some signal it expects, I guess. And if it doesn't get video, it assumes there's nothing there, so can't use it just for keyboard.

                                            Right now I've got 3 keyboards, two wireless mice, one monitor with a cable dangling and 3 cables from the 3 in-use boxen (so when I switch I'm only wearing out the end of a cable that's handy in reach, not a port on something valuable I have to do contortions to get at) which is a nuisance compared to a good KVM, for sure, but a KVM that only half works was worse.

                                            The guy fighting with the AMD64 *was* the repair shop, only one left for about an hour in any direction, not counting Worst Buy. It's a dying industry. Life was better when there was a clone shop on every corner, and everyone could afford to keep a broad inventory in stock. The Chinese mob put a lot of the orientals' shops out of business in California (and probably elsewhere), and disposables and phones are doing for the rest of it. I'm not even tripping over many knowledgeable amateurs anymore. PC user groups are a thing of the past, or populated by octogenarians. Here's hopin' you stay in business, long and profitable.

                                            I don't use the mouse for gaming, makes my wrist hurt, so I'm no help to ya there.

                                            --
                                            And there is no Alkibiades to come back and save us from ourselves.
                                            • (Score: 2) by Hairyfeet on Friday July 31 2015, @03:25AM

                                              by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Friday July 31 2015, @03:25AM (#216140) Journal

                                              It was one of the cheapies, I really hate spending $$$ on something that should be a simple build like a KVM. I mean its just a fricking switch that lies to a PC and says something is hooked when it isn't, why is that so damned hard to get right? I'll probably end up biting the bullet and just getting the $30 mouse as it'll end up cheaper than having to buy a new KVM switch AND a new keyboard AND a new mouse. If somebody wants me to work on their USB only PC? Well I got a couple spare 16x9 monitors in the back, I'll just slap one on.

                                              And the shops that died? Did so because they were DUMB, they tried to pretend the world doesn't change instead of changing with it! Now instead of the big sellers being office boxes its HTPCs, home theater installs, and service calls. You'd be surprised how many are willing to pay $40 an hour (plus $25 service call fee) just to have their PC worked on at home so they don't have to drag the thing in. Hell I have one customer that pays me at least every other month just to do dumb little shit like show him how to do some browser settings or clear his cache. He says "I'd rather just pay to have you come out so I know that it works where I'm at, rather than bring it in and then get it home and find I don't know how to get it going"...customer is always right LOL! I'm lucky in that a lot of my HTPC customers aren't snobs, as long as its a decent price and lets them watch YouTube and BS on FB from their big comfy chair? They couldn't care less if I built the thing in some no name black case I had laying around.

                                              You should probably try a little of that on the side bud, pretty much anything from a Pentium D on up makes for a decent HTPC as long as you optimize and install the right software. Got a customer downstairs with an old 2.6GHz Pentium D I just threw together out of the parts pile and all he does is rave to anybody who will listen about the thing, I showed him how to download concert videos from YouTube so he had me get him a 2TB and put in it and now he has every live show from AC/DC to The Wall just blasting on his 40 inch, showing it off to anybody who'll look. That is another mistake too many shops make, you have to get folks excited about what they are getting, show 'em the cool shit they can do. Folks are happy to shell out when they can show off their purchases, and they are happy to shell out when they feel you are treating 'em right. Show 'em how MediaPortal works, show 'em how to download a YouTube video, simple shit like that makes a big impression. The shop down the street charged HALF what I did...he is out of business, I'm still here, why? Because he always did the bare minimum, didn't install an AV, didn't install the patches, all he would do is remove the bug or fix the problem and hand it back with a bill, not even attempting to help them avoid that problem in the future. People get frustrated with PCs, make 'em feel at ease, make 'em feel like it could happen to anybody so they don't feel stupid, little things like that.

                                              Anyway if you have any luck with the USB KVM let me know. If your wrist hurts have you tried one of the "handshake" style mice? the kind you use in the handshake position? My former boss had RSI and used one of those, he could frankly kick your ass at Q3 Deathmatch and he could use that all day without hurting his wrist. I've been trying to use a big ass flightstick... its setting in the box because I can't get the hang of the damned thing LOL. I guess I've used KB and mouse so long trying to go back to stick is hell, I guess I'll have to dload one of the flightstick profiles from one of the YouTubers that are good at it, because damned if I can get the thing set up right. BTW sorry if there is errors in grammar or spelling, the little one had me up at the crack o dawn looking at furniture...groan. She hates living in an apt so we are fixing up a little 3 bedroom singlewide that belonged to my late sister. Managed to swap out nearly all the painting, carpentry, and plumbing, once its all done all we have to do is slap in the furniture and turn on the cable...man I am soooo not looking forward to having to move 10 years worth of fricking stuff LOL!

                                              But good luck on the search, need any tips or tweaks you know where to holler.

                                              --
                                              ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
                                              • (Score: 2) by Reziac on Friday July 31 2015, @04:14AM

                                                by Reziac (2489) on Friday July 31 2015, @04:14AM (#216161) Homepage

                                                Yeah, I know what you mean -- back when I was doing housecall PC repairs and custom builds, my customers were so loyal they'd wait forever rather than call someone else, mostly cuz I made 'em feel like they could use their shit how they wanted to use it, instead of leaving 'em with this mystery beige box they didn't understand. While back one of 'em sent me $300 out of the blue, cuz that's what she felt like she owed me for misc. service and help over the years. (Way more than I ever charged her.) How's that for tangible appreciation! Been out of that business a long time, tho, and haven't kept up on new tech.

                                                Having watched the clone industry fall apart around Los Angeles -- mostly it was the cheap disposable PCs, got real tough even at the low end to compete with $300 HPs from Worst Buy, and there was never enough volume of custom work to make up the difference. The survivors had networking contracts that paid the rent, cuz custom boxes and components for DIYers wasn't enough anymore.

                                                I suspect with a handshake mouse, my shoulder would be killin' me instead of my wrist. It ain't that hard to make a game be keyboard-friendly, and if they're not, well, I can't claim I'm all that interested in most of 'em anyway.

                                                Hey, I'd take the singlewide over an apartment any day, at least you don't have to share walls with your neighbor!

                                                --
                                                And there is no Alkibiades to come back and save us from ourselves.
                                                • (Score: 2) by Hairyfeet on Saturday August 01 2015, @07:10AM

                                                  by Hairyfeet (75) <{bassbeast1968} {at} {gmail.com}> on Saturday August 01 2015, @07:10AM (#216688) Journal

                                                  you sound like the wife, she has been over there painting at night just trying to rush it along a little faster. I don't know which has her more wound up, having a full kitchen with a dishwasher instead of a kitchenette or the fact she is getting one of the two spare bedrooms for her photography, the other is gonna be a little mini-studio for me. I've lived in apts most of my life so it really doesn't phase me, but losing my dad in May and mom having declining health means living closer would probably be wise and with the singlewide literally being next door if there is a problem i can just walk across the way and be there when I'm needed.

                                                  And it sounds like they did the same dumb shit in LA that many of 'em tried to do here, which was compete on price...NEVER a good idea, they will always have economies of scale on their side. Instead you compete with quality of service and offers they don't have like integrating with home theater setups and doing service calls. And you're right they are VERY loyal because they know I'm gonna make sure its doing everything they want it to do, if it takes 5 minutes or 5 hours. I always tell 'em "If you want a cheap pile of junk? Feel free to grab a 'Walmart Special', just don't be surprised when it drags thanks to all the crapware and 'service' is one guy in India that keeps telling ya to reboot". Hell I WISH there was a Worst Buy here, I had a shop down the street from a Worst Buy once, made a mint off their irate customers bringing their 'fixed' PCs to me LOL. But I got tired of the hour long commute through crazy traffic and came back home. Now the only "competition" I have is the phone place down the street and I get along with them fine, if somebody needs phone work I send them their way, if their customer asks about computers or home theaters they send 'em my way, works nicely.

                                                  But if you ever decide you want to make a little extra scratch just check out mediaportal, its crazy easy to set up, has hundreds of drop in plug ins so you can easily customize it to their tastes, and it'll run on pretty much any PC made in the last decade. You can take pretty much any black box, slap on MP and add a cheap $20 Lenovo wireless keyboard and voila! Instant HTPC. With every TV coming with HDMI its plug and play, just pop into the GPU control panel and set TV mode and you're done. You'd be surprised how many folks are sick of watching YouTube on their little screens but have no clue how to get it on their TVs and they quickly find all those little "sticks" like Roku just too limited. Once you've set up a couple HTPCs word of mouth takes care of the rest, easy peasy.

                                                  --
                                                  ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
      • (Score: 2) by Reziac on Sunday July 26 2015, @07:40PM

        by Reziac (2489) on Sunday July 26 2015, @07:40PM (#213952) Homepage

        Got the new POST tester (stuff is trickling in from Amazon practically one gadget at a time) and happy as hell with it. Initial check shows it's reporting accurately. Good documentation booklet too, has all the codes. It'll earn its keep.

        Now eyeing the CPU from the resurrected Dell and wondering if it's worth swapping onto the better mobo ... two different Intel quad cores.

        --
        And there is no Alkibiades to come back and save us from ourselves.