SoylentNews is people
SoylentNews
AMD Ryzen 7 Launches
Reviews and benchmarks for AMD's Ryzen 7 8-core desktop CPUs flooded out at 9 AM EST/6 AM PST:
Along with the new microarchitecture, Zen is the first CPU from AMD to be launched on GlobalFoundries' 14nm process, which is semi-licenced from Samsung. At a base overview, the process should offer 30% better efficiency over the 28nm HKMG (high-k metal gate) process used at TSMC for previous products. One of the issues facing AMD these past few years has been Intel's prowess in manufacturing, first at 22nm and then at 14nm - both using iterative FinFET generations. This gave an efficiency and die-size deficit to AMD through no real fault of their own: redesigning older Bulldozer-derived products for a smaller process is both difficult and gives a lot of waste, depending on how the microarchitecture as designed. Moving to GloFo' 14nm on FinFET, along with a new microarchitecture designed for this specific node, is one stepping stone to playing the game of high-end CPU performance.
Ryzen 5 chips will be released sometime in "Q2", and are presumed to have 4 to 6 cores with hyperthreading enabled. One of these has been revealed: the Ryzen 5 1600X. It has 6 cores, and equivalent clock (3.6 GHz) and turbo (4.0 GHz) speeds to the $500 8-core flagship Ryzen 7 1800X. Ryzen 3 chips will be released in the second half of the year, and include quad-cores with no hyperthreading.
The Intel Core i7-7700K maintains a lead in single-threaded performance, but the Ryzen 7 chips lead in many multi-threaded benchmarks (sometimes beating the $1089 Intel Core i7-6900K).
AMD Ryzen 7 Great, but Underwhelming for Gamers
PC Gamer has just published a lengthy and in-depth review of the new AMD Ryzen 7 CPU, (unfortunately broken down into seven separate pages). Their verdict? The AMD Ryzen 7 has plenty of power, but underwhelming video game performance.
The good news is that in heavily-threaded workloads, Ryzen looks every bit as potent as AMD has suggested. Processors with eight cores and sixteen threads have been relegated to the extreme performance community by Intel going back to the i7-5960X, with a starting price of $1000. AMD takes that ludicrous price tag and kicks it to the curb with a $500 halo part, and then ups the ante with $400 and $330 offerings. And just for good measure, every single Ryzen processor is multiplier unlocked, so enthusiasts can coax even more performance out of the parts.
If you've been longing to try out a beastly 16-threaded monster, the price barrier to entry just got slashed from around $2000 for a complete high-end build to around $1250. Ryzen also easily beats the old FX-series in every test I ran, with even the entry-level Ryzen 7 1700 at times more than doubling the performance of the FX-8370. Yeah.
The bad news is that Ryzen's single-threaded performance, and perhaps more importantly its clock speeds, aren't quite as impressive. In some workloads, Ryzen does a great job at keeping up with Intel's Haswell/Broadwell architectures, but it's about 5-10 percent slower per clock compared to Skylake/Kaby Lake. Toss in the higher clocks of Kaby Lake and 10 percent can quickly turn into a 40 percent deficit in some tests. Among the use cases where Intel maintains a decent lead over Ryzen is gaming performance, though this varies by game.
The result is some great wins for Ryzen, and some painful losses. But the losses are actually a bit odd. To be frank, Ryzen and the AM4 platform feel a bit undercooked right now.
It appears that Intel will remain the choice of CPU for video gamers, at least until the Zen+.
AnandTech's Interview with AMD CEO Lisa Su
http://www.anandtech.com/show/11177/making-amd-tick-a-very-zen-interview-with-dr-lisa-su-ceo
Q5: How vital was it to support Simultaneous Multi Threading?
LS: I think it was very important. I think it was very complicated! Our goal was to have a very balanced architecture. We wanted high single threaded performance, and SMT was important given where the competition is. We didn't want to apologize for anything with Zen – we wanted high single thread, we wanted many cores, but sorry we don't have SMT? We didn't want to say that, we wanted to be ambitious and give ourselves the time to get it done.
[...] Q8: Do you find that OEMs that haven't worked with AMD are suddenly coming on board?
LS: I will say that we have engagements with every OEM now on the high-performance space. Twelve months ago, a number of them would have said that they don't have the resources to do multiple platforms. So yes, I think momentum helps in this space.
Q9: At Intel's recent Investor Day we learned that future chips will incorporate multiple dies on the same package. This allows a semiconductor firm to focus on smaller chips and potentially better yields at the expense of some latency. Given what we predict will happen, what is your opinion on having large 600mm2 silicon? Is there a future?
LS: There has been a lot of debate on this topic. I find it a very interesting debate. Certainly on the graphics side we view High Bandwidth Memory (HBM) and the ability to get that interconnect between the GPU and memory to be extremely differentiating. So certainly we will use that throughout our graphics roadmap. If you look at our CPU roadmap, I do think there's a case to be made for multi-chip modules. It depends on the trade-offs you have to do, the bandwidth requirements you have, but yes as the process technology becomes more complicated, breaking up the tasks does make sense.
The day of reckoning.
Also at Tom's Hardware (which includes gaming benchmarks, unlike the AnandTech review), Ars Technica ("AMD Ryzen is an excellent workstation CPU—shame its gaming performance is weak"), PCWorld, HEXUS.
Previously: New Details of AMD's Desktop Zen/"Ryzen" Chips Released
AMD Ryzen Processor Hype is Building
Original Submission #1 Original Submission #2 Original Submission #3
(Score: 3, Informative) by Marand on Monday March 06 2017, @02:22AM
[First off, sorry for the delayed response. I saw the response notification while on mobile and forgot about it by the time I had a chance to sit down and respond. Just remembered it again...]
If cost is no concern at all, probably Intel, at least for the moment. Even if Ryzen's better in every way it's going to take time for software to catch up. As for future-proofing, there's no way to say. AMD's last architecture bet heavily on certain types of workloads and that didn't pan out, while Intel's plan did, so they've been ahead for a while. Prior to that, Intel bet on longer pipelines with Pentium 4 around when AMD went for 64bit, and AMD won that one.
If it were my call, I'd go for Ryzen right now and hope for the best, but that's as much because I don't like Intel's business practices as any other reason. AMD seems to have a good enough design this time that I feel like I can back the company that deserves it while still getting good performance, and a good price/performance value.
Specifically I don't know how much of my typical workload is multithreaded.
It's nice when software can use the cores directly, of course, because you can see a direct impact (faster compiles/renders/etc.) but it's not the only way your system can use them. Even if your software can't take advantage of it, you can still benefit from extra cores as long as your OS can, because it can put different processes on different cores and keep things running smoothly. Like the example I gave of encoding video for streaming while gaming, or being able to dedicate cores to virtual machines without sacrificing performance on the host OS, etc.
Sorry I'm not giving you a clean, clear answer here. There isn't one to give, it's a muddy subject with no single correct answer. Most anything will be "good enough" for most tasks, and picking something optimal to a specific workload has as many answers as workloads.
It also doesn't help that it's hard to trust benchmarks and other information, which makes it even harder to give an answer. Intel has repeatedly been caught using its compiler (ICC), which has been super-popular especially on Windows, to fudge benchmarks or cripple competitors' chips. They had it check the CPU vendor information and disable optimisations for non-Intel parts in the past, and more recently attempted to fudge a popular Android benchmark by having the compiler check if the benchmark was running and, if so, replaced certain calculations with no-ops on Atom CPUs to make them score better. So, even when Intel really is performing better, it's hard to say how much of it is real, and how much of it is due to compiler shenanigans.