SoylentNews is people
SoylentNews
Some Soylentils were disappointed by the gaming performance of AMD's Ryzen CPUs when they were launched last month. By now, updates have eliminated some of the advantage that Intel CPUs had, but the potential gains differ depending on the game:
The first big Ryzen patch was for Ashes of the Singularity. Ryzen's performance in Ashes was arguably one of the more surprising findings in the initial benchmarking. The game has been widely used as a kind of showcase for the advantages of DirectX 12 and the multithreaded scaling that it shows. We spoke to the game's developers, and they told us that its engine splits up the work it has to do between multiple cores automatically.
In general, the Ryzen 1800X performed at about the same level as Intel's Broadwell-E 6900K. Both parts are 8-core, 16-thread chips, and while Broadwell-E has a modest instructions-per-cycle advantage in most workloads, Ryzen's higher clock speed is enough to make up for that deficit. But in Ashes of the Singularity under DirectX 12, the 6900K had average frame rates about 25 percent better than the AMD chip.
In late March, Oxide/Stardock released a Ryzen performance update for Ashes, and it has gone a long way toward closing that gap. PC Perspective tested the update, and depending on graphics settings and memory clock speeds, Ryzen's average frame rate went up by between 17 and 31 percent. The 1800X still trails the 6900K, but now the gap is about 9 percent, or even less with overclocked memory (but we'll talk more about memory later on).
(Score: 2) by Hairyfeet on Tuesday April 11 2017, @10:14PM (1 child)
Actually I have to wonder how much of "bulldozer not paying off" was due to the Intel Cripple Compiler [theinquirer.net] because I have an FX-8320e and use a lot of GCC compiled programs like Audacity and even on super heavy workloads like applying audio effects across multiple tracks the performance is just superb, and many of the games I play (like War thunder) are made by overseas developers that likewise don't use ICC and I'm getting 90+ FPS while having everything cranked up to movie settings. And considering I only paid a hair over $600 for the system with 5Tb of storage with 240gb SSD for boot, R9 280, 16Gb of RAM, gamer board with quad SLI/CF support and a BR drive for data backups? I sure as hell ain't complaining and could easily see myself using this system through 2020 and even possibly later.
But the whole "ZOMFG its not beating teh Intel on gamez ZOMFG!" smells like a little bit of Intel PR spin to me, especially considering we saw in the antitrust transcript that Intel had employed that tactic before with their "advertising partnerships" where review sites got million dollar checks to advertise Intel chips as long as they sang the Intel gospel. I mean we are talking about a chip that gets within 25% of an $1100 CPU at less than half the price why the fuck isn't every site running with that as the headline? You can literally build an ENTIRE Ryzen system for less than the price of the Intel chip alone and you are getting 75%+ of the performance? Who doesn't think that is just insanely good, especially when Ryzen doesn't have ANY of the important features removed like ECC, OCing ability, and support for hardware VMs unlike what Intel does?
Like I said Ryzen 7 is complete overkill for games but I can see why gamers are snatching them up, I mean 16 threads with 75% of the performance of an $1100 chip for just $500? That is just nuts and I'm sure Ryzen will take the place of the FX-8 as the chip for streamers and those of us who record our gameplay footage but the fact that the mainstream press is completely ignoring all the positives to focus on a single negative that with any thought shows isn't really a negative given the price? The whole thing just smells funny to me.
ACs are never seen so don't bother. Always ready to show SJWs for the racists they are.
(Score: 2) by Marand on Wednesday April 12 2017, @12:20AM
Actually I have to wonder how much of "bulldozer not paying off" was due to the Intel Cripple Compiler
I'm sure that didn't help, but I'm referring to how AMD gambled on integer-heavy workloads by having each "module" essentially be two cores for integer work but only one for floating point. That didn't pay off in some cases, and on top of that, they also gambled on multi-core workloads but games were really slow to make the switch. The architecture is nearly six years old now, and most games still don't use more than four at best. The multi-core bet is finally starting to pay off (albeit too late for bulldozer at this point), but the int/fp one never really did.
As for the whole ICC bullshit, it's one of the things I had in mind when I said I dislike Intel's business practices. The FTC busting them for it didn't even change anything, because all they did was add a disclaimer on the site stating that ICC might not be generating good executables for non-Intel CPUs. Since then the same shit happened again against ARM [theregister.co.uk], this time by having ICC skip certain instructions used in a benchmark when the CPU was an Atom one only.
But the whole "ZOMFG its not beating teh Intel on gamez ZOMFG!" smells like a little bit of Intel PR spin to me, especially considering we saw in the antitrust transcript that Intel had employed that tactic before
Could be shilling, tribalism, or just people getting caught up in "LONGER BARS IS BETTERS!" dick-waving. Any are likely.
One thing I've noticed is that the people ragging on Ryzen performance in games are conveniently ignoring that, even when Ryzen loses in average or maximum FPS, it tends to have fewer lows as well. That's not as brag-worthy as throwing around bigger numbers, but can be a smoother experience overall and shouldn't be ignored. A few reviews have mentioned it, at least, so it hasn't been completely left out.
I mean we are talking about a chip that gets within 25% of an $1100 CPU at less than half the price why the fuck isn't every site running with that as the headline?
I'd wager they don't put that directly in the title because then you'd have no reason to read the rest of the article. To be a fair, most of the reviews I've seen so far have essentially stated that as the final conclusion, and I remember one even saying that buying a Ryzen 7 is basically like getting the $1000 Intel chip plus a really nice GPU for free. (e.g. "Buy the Ryzen 7 and a nice GPU instead.")
Overclocking wasn't touched on much in the initial reviews, but it's been getting a lot more attention since, and I've even seen it come up in a couple R5 reviews. I think the general consensus with the R7s, after the initial day-1 reviews, has been that the 1700 is the best bang-for-your-buck because it's the lowest price point and still generally overclockable to ~4ghz, even with the Wraith Spire fan that comes with it.
I didn't even know about the OCing when I chose the 1700, I was just interested in it for the lower TDP and figured that, based on the benchmarks I'd seen, its performance was close enough to the 1700x and 1800x that I wouldn't feel like I'd be missing out. Finding out that I can OC it well if I get the itch was just a nice post-purchase bonus. :D
but the fact that the mainstream press is completely ignoring all the positives to focus on a single negative that with any thought shows isn't really a negative given the price? The whole thing just smells funny to me.
I've seen a couple reviews focus heavily on that, but I've seen just as many have either argued that it's mostly irrelevant because everything tested is fast enough, or that it's a good trade-off for most people. Even the ones that emphasise the Ryzen gaming benchmarks and treat it negatively seem to largely be claiming that Ryzen is a return to competitiveness for AMD. Most of the damage control seems to be coming from the comment sections, where a handful of people focus entirely on the gaming benchmarks and nothing else.
Repeating what I said a few paragraphs ago, this could just as likely be tribalism or people getting too hung up on numbers with no regard for context. That's a trap you can see reviewers fall into often because they review so many products, usually mostly good, and end up nitpicking over things that won't matter most of the time as a way to differentiate the products. Smartphone reviews have had a problem with this for a while, for example, where an otherwise good phone gets slagged for having a slightly worse camera than some other model, or only having a 400ppi screen instead of a 420ppi one, etc.