Stories
Slash Boxes
Comments

SoylentNews is people

posted by martyb on Friday July 28 2017, @08:12AM   Printer-friendly
from the has-no-internal-GPU dept.

AMD has launched the Ryzen 3 1300X and Ryzen 3 1200. Both models are quad-core CPUs with only one thread per core, a TDP of 65W, and both come bundled with a "Wraith Stealth" cooler. The 1300X ($129) has the same clocks as the Ryzen 5 1500X ($189), but with no simultaneous multithreading and half of the L3 cache. The Ryzen 3 1200 ($109) has lower clock speeds.

Like all of the Ryzen CPUs released so far, integrated graphics is not included, but both are unlocked processors that can be overclocked.

There are two ways to approach analyzing the competition: configuration and price. For configuration, Ryzen 3 are quad-core CPUs without simultaneous multithreading, which would put them up against the Core i5 CPUs, which range from $182 to $239. Comparing on price, the Ryzen 3 1300X at $129 fits between the Core i3-7100 ($109) and Core i3-7300 ($149).

The conclusion of the AnandTech review does not take into account overclocking, which will be done in a later article. Tom's Hardware reviewed just the Ryzen 3 1300X, overclocking it to 3.9 GHz easily. After the overclock, the 1300X pulls ahead of Intel's Core i3-7100 and i3-7300 in gaming benchmarks.


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This discussion has been archived.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.
(1)
  • (Score: 2) by RamiK on Friday July 28 2017, @11:06AM (2 children)

    by RamiK (1813) on Friday July 28 2017, @11:06AM (#545715)
    • (Score: 2) by takyon on Friday July 28 2017, @11:58AM

      by takyon (881) <takyonNO@SPAMsoylentnews.org> on Friday July 28 2017, @11:58AM (#545726) Journal
    • (Score: 0) by Anonymous Coward on Friday July 28 2017, @07:59PM

      by Anonymous Coward on Friday July 28 2017, @07:59PM (#545953)

      "Su added, however, that AMD isn't thinking of mining as a long-term problem. "We didn't have cryptocurrency in our forecast, and we're not looking at it as a long-term growth driver, though we'll continue to watch the blockchain technologies as they go forward," Su said. "

      1.) "We didn't have cryptocurrency in our forecast"

      why the hell not? this happened at least once already! What's your problem? Sounds like BS, like they think that all their end users are dumb windows gamers or something...

      2) "and we're not looking at it as a long-term growth driver, though we'll continue to watch the blockchain technologies as they go forward,"

      so, you're going to "keep an eye on it" but with more presumptions, improving the odds that you waste vast potential next time too? Why not just make a plan that works for various scenarios? up front. not wait and then react, like a slow, big, dumb ass, company?

      dial back the Smug, AMD. you're not as smart as you think you are. If you were, you would have used this "one/second time miner craze" to gain advantage with GF and improving your bottom line even more at the same time. Instead, retailers are out of stock everywhere. That should never, ever happen in 2017. We have these things called computors and they can decide when to order, ramp up production, improve efficiency, etc. There may be a wikipedia page you can reference. I would blame the card manufacturers, distribution networks and retailers for their outdated processes but they can't even get GPUs from the people that supposedly make GPUs! You still don't have shit on the shelves and this has been going on for months. It's just sad, really. I like AMD compared to other chip makers.

      hurry up and get your shit together!

      Hmmm, maybe AMD is purposely sabotaging cryptocurrency on behalf of other interests? Sure is a lot of money being left on the table for incompetence to explain it...

  • (Score: 2, Insightful) by engblom on Friday July 28 2017, @02:46PM (10 children)

    by engblom (556) on Friday July 28 2017, @02:46PM (#545774)

    Am I old fashioned, but I found the last comment about not taking overclocking into account to be stupid and unnecessary. Normal intelligent people do not overclock. They run the CPU at the speed the manufacturer considered it safe to run them. If AMD could boost their performance by just raising the clock frequency, would they not do it?

    Of course CPUs should be compared without overclocking. Most people buy big brands computers with the CPU ready installed. Those ricers building overclocked computers form an ultra tiny fraction. I bet they are in far minority even among the computer builders. I have built many, many computers but never overclocked a computer and I have not had to gamble with stability.

    • (Score: 0) by Anonymous Coward on Friday July 28 2017, @03:05PM (1 child)

      by Anonymous Coward on Friday July 28 2017, @03:05PM (#545779)

      Am I old fashioned, but I found the last comment about not taking overclocking into account to be stupid and unnecessary.

      You may or may not be old fashioned, but I consider the information that those are results that were achieved without overclocking worthwhile. It means that I can rely on those results to give me a proper estimate of what to expect from that processor. With Tom's Hardware, OTOH, I'd have to carefully look at all data to check whether they include results from overclocked processors or not.

      OTOH, I don't see why gamers shouldn't overclock their gaming machines if they are not using them for serious work. If their computer crashes, well, the worst thing that happens is a lost chance at winning a game.

      If AMD could boost their performance by just raising the clock frequency, would they not do it?

      Because whoever buys the processor might use it for productive purposes, and will rightly expect it to work as flawlessly as possible.

      Basically by allowing overclocking, AMD effectively is selling in the same product both a reliable but lower performing CPU for serious work, and a faster but potentially unstable CPU for gaming. And if you are stupid enough to overclock a processor for productive work and it crashes, AMD is safe from liability because you operated it out of specification.

      • (Score: 2) by takyon on Friday July 28 2017, @03:40PM

        by takyon (881) <takyonNO@SPAMsoylentnews.org> on Friday July 28 2017, @03:40PM (#545808) Journal

        If you look at the graphs in the Tom's review, you can see that the overclocked CPU is compared to the same CPU run at the stock clock. So you are getting both results in each graph. The overclocked chips are labeled "@ 3.9" in this case.

        --
        [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
    • (Score: 0) by Anonymous Coward on Friday July 28 2017, @03:06PM

      by Anonymous Coward on Friday July 28 2017, @03:06PM (#545780)

      Overclocking is at least 'on the table' if not a desired feature.

      Personally I test overclocking on new processors I buy, find the highest stable point which doesn't throw errors (often anywhere from 10-30 percent more than the rated clock rate), run some burn in tests, then use the power management features to run it at stock clocks, with some available overhead if future applications require it.

      Since at least the pentium 4 era the clock rate was usually *NOT* the determining factor in what a processor could be run at. More often than not it was the maximum safe junction temperature before the part became erratic, or the maximum TDP of either included, or available heatsinks. If you had a heatsink capable of dissipating 300-600W of waste heat, and a motherboard capable of supplying said chip power, you could running much higher clock rates than current desktop processors allow on essentially the same silicon. If you are in doubt, go look at the clock rates for Oracle/Sun, IBM, etc. Many of them are running 20+ percent higher clocks than Intel, due solely to the ability to dissipate higher amounts of heat per chip. The past few generations of Intel/AMD chips have shown those higher clocks can be attained even on consumer hardware given sufficient cooling resources, and even get to speeds that previously were thought to require cryonic cooling methods, instead only utilizing water or air cooling devices.

    • (Score: 2) by etherscythe on Friday July 28 2017, @03:29PM

      by etherscythe (937) on Friday July 28 2017, @03:29PM (#545803) Journal

      Ryzen chips (at least the X SKUs) have an "eXtended Frequency Range" feature somewhat comparable to Intel's Turbo Boost, which is essentially an automatic overclock feature. The better your cooling, the better the Ryzen chip will perform with no user tweaking required. You can overlook this capability, but to do so is to perhaps leave some performance ROI on the table.

      --
      "Fake News: anything reported outside of my own personally chosen echo chamber"
    • (Score: 3, Informative) by takyon on Friday July 28 2017, @03:36PM (3 children)

      by takyon (881) <takyonNO@SPAMsoylentnews.org> on Friday July 28 2017, @03:36PM (#545807) Journal

      It is trivial to overclock Ryzen using AMD's own utility [amd.com] and the BUNDLED cooler. Your ricer comment is stupid and unnecessary. Everyone who buys the CPU gets the cooler, and overclocking the thing is almost no work at all.

      AMD sells them at a certain frequency due to binning, as some chips can run hotter than others. Stock frequencies are to ensure a baseline of performance while maintaining a quiet system with less power draw [wikipedia.org].

      All of the Ryzen chips have unlocked multipliers, while most of the Intel Core i3 chips do not. The AMD chips are designed to be overclocked [overclockers.com]. Today's overclocked CPUs are more stable than the CPUs you used in the 90s.

      --
      [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
      • (Score: 1) by nitehawk214 on Friday July 28 2017, @08:29PM (2 children)

        by nitehawk214 (1304) on Friday July 28 2017, @08:29PM (#545963)

        The stock coolers are quite good these days. The only reason I don't use one is because I wanted a lower rpm / lower noise cooler, not a more effective one.

        --
        "Don't you ever miss the days when you used to be nostalgic?" -Loiosh
        • (Score: 2) by takyon on Saturday July 29 2017, @12:36AM (1 child)

          by takyon (881) <takyonNO@SPAMsoylentnews.org> on Saturday July 29 2017, @12:36AM (#546063) Journal

          Click the cooler link. "Wraith Stealth" is apparently a quiet fan. Not sure if it is quiet enough for your needs, but there are some decibel amounts listed in that article. They also test the two bigger ones but not the Wraith Stealth in that review.

          --
          [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
          • (Score: 1) by nitehawk214 on Saturday July 29 2017, @04:11PM

            by nitehawk214 (1304) on Saturday July 29 2017, @04:11PM (#546297)

            I need a new computer sooner than later, I may check this out. The one I have currently is nice and quiet, but too goddamn big.

            --
            "Don't you ever miss the days when you used to be nostalgic?" -Loiosh
    • (Score: 2) by tekk on Friday July 28 2017, @04:49PM

      by tekk (5704) Subscriber Badge on Friday July 28 2017, @04:49PM (#545857)

      You say that, but some people may be overclocking without knowing it. As a standard feature on my motherboard, it will actually overclock my CPU slightly if it needs it and the thermal situation is looking okay.

    • (Score: 2) by linkdude64 on Friday July 28 2017, @10:37PM

      by linkdude64 (5482) on Friday July 28 2017, @10:37PM (#546015)

      "Normal intelligent people do not overclock."

      I would probably wager the IQ of people who overclock would be 1 or 2 points above average. It requires some passion for technicality which would likely incline the average to fields that require critical thought. Abnormal? Sure, but why is that a bad thing?

(1)