Discussion AMD Ryzen MegaThread! FAQ and Resources

Page 82 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Recent (August 14) IPC comparison
https://cpugrade.com/articles/cinebench-r15-ipc-comparison-graphs/

At 3.00 GHz, the past three generations of Intel microprocessors suddenly find themselves very close to each other, while AMD's Zen architecture squeezes between Broadwell and Skylake.
Click to expand...

cbr15-ipc-comparison-3ghz-scores.png
 
That 139 -> 146 is a 4.79% increase. If only I would've always thought about these single digit jumps being significant in close quarter scenarios... Oh wait. I actually did know that.

So we can say that is now at Broadwell-E level and just a tad behind Skylake and Skylake-E, right?

Cheers!
 
It seems as if the IPC of RyZen increases for minutes! The PcPer review at launch put Ryzen about 8% behind Broadwell. The recent guru3d article put is on pair with Broadwell. And the latest cpugrade article puts it about 2% above Broadwell.

Except there is no real increase in IPC, simply the illusion that the IPC is increasing by cherry picking a pair of reviews in the appropriate order (first PcPer, then recent guru3d, then cpugrade) and then ignoring every other review that draws a different picture. Those cherry picked reviews have AMD chips running faster and Intel chips running slower than rest of reviews. What causality! Moreover, I find it weird that some recent reviews are using dirty tricks, such as measuring performance on overclocked AMD chips but reporting them as "stock".

Take reviews from the same site. Consider guru3d. In the initial review of Zen, at launch, they measured 156 and 1618

http://www.guru3d.com/articles_pages/amd_ryzen_7_1800x_processor_review,10.html

In the recent review guru3d measures 161 and 1637.

http://www.guru3d.com/articles_pages/amd_ryzen_threadripper_1950x_review,10.html

Therefore, the ST score increased by 3% and the MT score increased by 1%. The average of both 2%. It is evident that the IPC has not increased by 10%. There is no magic patches or updates for Zen.

RyZen IPC continues being a good 10--20% behind Broadwell, depending of the workload, one can check this when one consider good reviews or an average of good reviews.
 
Processors FAQ
Q How does IPC compare between AMD's Ryzen and Intel's Kaby Lake?
A Ryzen’s IPC nearly matches Kaby Lake but at worst lags by 9%.

What is IPC and how is it calculated?
"IPC (Instructions per cycle) refers to the number of calculations a CPU can perform per clock cycle. If two CPUs have the same IPC they will have the same performance at the same frequency. For the practical purpose of comparing performance between different CPU architectures a workload that represents “typical use” is tested on both CPUs at the same clock speed to produce an IPC figure per core. IPC is measured under multi-core workloads to ensure that the resulting figures reflect total processing throughput. When the relative IPC between architectures is known, it simplifies comparisons between them.
Turbo Boost.
Nearly all modern CPUs have the ability to turbo boost. Historically turbo boost would only engage whilst some cores were idle but on recent architectures, including both Skylake and Kaby Lake, turbo boost can engage even whilst all cores are active. This obfuscates the meaning of base clock which can no longer be used for meaningful comparisons. For example the Skylake Core i5-6400 has a base clock of 2.7 GHz but all four cores are able to turbo to 3.1 GHz which gives it an effective base clock of 3.1 GHz. We don’t yet know under what conditions and to what extent Ryzen chips are able to turbo boost.
Skylake and Kaby Lake have the same IPC.
Taking the Kaby Lake i5-7500 which has a quad core boost frequency of 3.6 GHz and the Skylake i5-6600 which also has a quad core boost frequency of 3.6 GHz, then comparing the quad core mixed performance shows that they score 452 and 451 respectively. This demonstrates near identical IPC between Kaby Lake and Skylake (averaged over 11,000+ samples).
Kaby Lake vs Ryzen IPC estimate (Feb 27, 2017). (Note the Date!)
Until we know the exact frequency profile of Ryzen under various loads we can only estimate its IPC.
Assuming the Ryzen 1700X operates at 3.4 GHz under quad core load (possible scenario):
(Ryzen 1700X = 416) vs (6700T with 3.4 GHZ boost = 418) implies Ryzen IPC equals Kaby Lake.
Assuming the Ryzen 1700X operates at 3.6 GHz under quad core load (less likely scenario):
(Ryzen 1700X = 416) vs (7500 = 452) implies Ryzen's IPC lags Kaby Lake's by 9%.
Assuming the Ryzen 1700X operates at 3.5 GHz under quad core load (most likely scenario):
Extrapolating from the above two results implies that Ryzen's IPC lags Kaby Lake's by 4.6%.
Conclusion.
Based on the small handful of samples we have seen, Ryzen's IPC could match Kaby Lake but at worst lags by 9%. The big unknown is Ryzen's overclockability as this will determine the actual performance levels delivered. The Kaby Lake 7700K can comfortably hold 4.8 GHz so, under quad core workloads the 1700X would have to overclock to 4.8 GHz or better in order to beat it. On the other hand, on an eight core workload assuming a clock speed of 3.4 GHz (stock) for the 1700X and 4.8 GHz (overclocked) for the 7700K, the 1700X wins by a whopping 41%. Even though these preliminary results are based on a limited number of samples one thing is for sure: Christmas has come early for workstation users."

http://cpu.userbenchmark.com/Faq/How-does-IPC-compare-between-AMDs-Ryzen-and-Intels-Kaby-Lake/108
 


On average, Zen IPC is 10--20% behind Kabylake. About 10% on throughput workloads and about 20% on latency workloads. And that is only on generic x86 code. The gap is bigger when we consider AVX code, because Kabylake has 2x256bit FMAC units, whereas Zen has 2x128bit.

The claim that Ryzen "at worst lags by 9%" is obviously false. Next a workload where RyZen IPC is 24% behind Kabylake

clock-audacity.png
 
Keep this in mind with LAME: v3.99.5 (February 2012). That is the *latest* release. And Audacity doesn't handle threading very well (I use it often myself) and uses SSE2 instead of AVX (which, ironically, is more predominant in games nowadays).

I wouldn't use Audacity+LAME as a good indicator of performance, although it is a necessary benchmark I guess. They need an update to keep up with times, lol.

Cheers!
 
As goldstone77 mentioned, the right way of comparing IPC is testing the processors at same clock frequency. So far I have only seen 3 reviews that compare single thread performance on cinebench at same frequency between broadwell, ryzen and skylake. Those reviews are the ones posted above: one in early february by PcPer and two in mid august by guru3d and cpugrade.

The two reviews in august give nearly identical IPC for Ryzen:
Guru3D 146
cpugrade 124
if we make the math 146/3,5*3 = 125 less than 1% difference in IPC
On the other hand the one in February shows about 5% lower IPC

Possible explanations:
a) All three are good reviews. And the differences between august and February are due to:
- Patches (no magic involved they can easily contribute 1-2% extra)
- More mature motherboards. Motherboard alone can have a significant difference for single thread cinebench.
b) one review is wrong the other two are right
c) two reviews are wrong and one is right :ouch:

Note: If anyone knows of other reviews that compare single thread performance on cinebench between broadwell, ryzen and skylake at same frequency please post them here. I would really like to see them.
 
Review-chart-template-2017-final.003-1440x1080.png

-8.5% 1500X@3.5GHz vs 7700k@3.5GHz
Review-chart-template-2017-final.002-1440x1080.png

+8.2% 1500X@3.5GHz vs 7700k@3.5GHz
Review-chart-template-2017-final.009-1440x1080.png

+4.3% 1500X@3.5GHz vs 7700k@3.5GHz in Physics showing Ryzens superior number crunching vs. Intel
-5.3% 1500X@3.5GHz vs 7700k@3.5GHz overall
But look at the 7600k with a base clock of 3.8GHz, which is 8.5% faster than the 1500X
+30.9% 1500X@3.5GHz vs 7600K@3.8GHz
-3.4% 1500X@3.5GHZ vs 7600K@3.8GHz
Review-chart-template-2017-final.012-1440x1080.png

-5.2% 1500X@3.5GHz vs 7700K@3.5GHz overall
But look at the 7600k with a base clock of 3.8GHz, which is 8.5% faster than the 1500X
0.2% 1500X@3.5GHz vs 7600K@3.8GHz overall

Intel is lucky Ryzen was on Samsungs inferior 14nmLPP, and IBM 7nmLP(Lead Performance not low power) is going to put a hurting on Intel!

Edit: changing 7.8% to 8.5% math error
 


Yeah it looks like AMD are doing pretty well with Ryzen and ThreadRipper.

I must admit I caved and got myself an R5 1600 rig (first totally new machine I've built in years). There's been enough time for prices to settle and bioses to get sorted out. I've not run into any issues with memory or anything else for that matter- put it all together, turned it on, installed Windows 10- done. Really easy build.

So far everything has been running very well. I haven't overclocked anything either, the 1600 at stock speed is pretty quick. For comparison I have a Haswell generation Xeon workstation at work (4 core / 8 thread), an Skylake Lappy (4 core / 8 thread) and the Ryzen build- it's the quickest of the 3 in everything I've tested- Ryzen might be behind the very highest clocked Intel chips like the 7700k in single thread stuff but It is no slouch. As for the machine it replaced- well the difference is huge (unsurprising given how old the FX 8320 is these days).
 


Not terrible, but definitely on the "cheap" side of gaming cases. (the blue LEDs scream cheap case).
Looks like it may have good... ventilation... but it's all at the bottom the case.... so it'll probably get hot... with that single exhaust fan in the rear.
 


If that thing has 4 USB3.x at the front, I love it. I don't care if it looks like a dumpster!

Cheers!
 
I agree Yuka, having like having all that connectivity in the up top in the front! But like James Madison says, it's very Dell like in construction! The LEDs I think are going to end up being a plus for marketing to the mainstream community.
 


You got me on that one! But it's not bad like it sounds.

Edit: I thought to myself now he is going to say something good about Ryzen as to show he is not bias. Nope that didn't happen! Fooled me!
 


Funny anecdote: when I first came to the UK and was touring the Experian data centers, I asked them about heating the water of the pipes used in water cooling.

Their reply was: "because no one has ever thought of doing it!".

The Mainframe racks produce quite a lot of heat that just goes to waste and is very inefficient when they could try and do something with that energy. Looks like someone somewhere out there found a way to make it work.

Cheers!
 


Actually, if you read the article, it's very good news for AMD. And it's even more positive if you go to the original source.
https://blog.qarnot.com/1500-amd-ryzen-pro-will-heat-homes-and-offices-next-year-in-bordeaux-france/

In our testing, based on 3D rendering computations, we saw a performance gain of 30 to 45% compared to the Intel i7
Click to expand...
Security is crucial for us.
.....
Ryzen PRO’s Transparent Secure Memory Encryption feature that enables OS independent memory encryption is therefore very interesting to provide additional security layers to our solution.
Click to expand...
Every processor heats. What matters is the Thermal Design Power, and Ryzen Pro is producing the same heat as the equivalent Intel CPUs we were using while providing twice as many cores. A good fit for our customers that use core count to evaluate their computing needs.
Click to expand...


 


Yes, Zen processors shine in GPU tasks.
 


They also shine in lots of CPU tasks...as benchmarks would indicate.
 
[video="https://www.youtube.com/watch?v=t8K2yc11eC4&ab_channel=HardwareUnboxed"][/video]
Ryzen 5 1600 vs. Core i7-7700K using Vega 64 & GTX 1080
Hardware Unboxed
Published on Sep 15, 2017

Core i7 vs. Ryzen 5 with Vega 64 & GTX 1080
Clash of Titans
By Steven Walton on September 15, 2017
1L8F3Y6.png

Side-by-side Breakdown of the Results
Not that long ago I compared the overclocked R5 1600 and i7-7700K in 30 games using the GTX 1080 Ti and at 1080p the Ryzen CPU was on average 9% slower. Here we see with the slower GTX 1080 the Ryzen CPU was just 5% slower, so that's pretty well in line with previous findings.
What's interesting to note about this side-by-side game comparison is that in the more modern and well put together titles, Ryzen is extremely competitive. In fact, the only real head scratchers here are Hitman, Dawn of War III and Total War Warhammer.
Click to expand...
1080_1080.png

In many other games such as Dirt 4, Doom, Sniper Elite 4, Battlefield 1, Rise of the Tomb Raider, The Division, Prey, Overwatch and Resident Evil 7 we found Ryzen to be extremely competitive. These results are of course based on the GTX 1080 handling the rendering work, so let's see how things look at 1080p with Vega 64.
Click to expand...
Vega_1080.png

Dropping in Vega 64 we see that overall Ryzen is actually slightly slower as it now trailed the 7700K by a 7% margin overall. Although the Ryzen 5 1600 processor does much better in Civilization VI, it now struggles in quite a few more titles than what was seen previously with the GeForce graphics card and we'll take a closer look at that in a moment.
For now let's see how things change at 1440p.
Click to expand...
1080_1440.png

At 1440p we are more GPU limited but even so we saw some strange things when comparing the Core i7 and Ryzen CPUs. A number of times Ryzen was at best able to match the 7700K at 1080p though at 1440p delivered noticeably better results. Only in Hitman Ryzen is slower by a 10% margin or greater and as a result is now just 2% slower overall.
Click to expand...
Vega_14.png

Moving to the Vega 64 results at 1440p, we again find that overall Ryzen was a mere 1% slower than the Core i7 processor. The margins on a per-game basis though are significantly different to what we just saw with the GTX 1080 so let's explore that a little closer.
I guess one takeaway here is that it's bad to generalize. For example claiming that Nvidia's DX12 performance handicaps Ryzen is certainly not true in all titles, though we might start to see more of this as newer games take better advantage of modern PC hardware.
For now though, Ryzen isn't always superior in DX12 titles and we can look to Hitman as an example. The 7700K is miles better in this game. Vega 64 doesn't always perform better with Ryzen either, as seen in titles like Dawn of War III, F1 2016 and Rise of the Tomb Raider where we witnessed the R5 1600 doing much better with the Nvidia GeForce GPU.
We also saw how much more the higher 1440p resolution brings both Ryzen and Vega into play. Ryzen still did well at 1080p for the most part, though Vega certainly appears much more competitive at 1440p versus 1080p.
Overall, the higher-end Vega 64 parts don't offer a great value while the complete opposite is true for Ryzen. If I didn't have money to burn, which I don't, and I was building a gaming system today intending it to last for the next three, four or possibly even five years, then I'd invest in the Ryzen 5 1600 rather than the more expensive Core i7-7700K, especially if $500+ GPUs aren't in your future.
Bottom line, it's safe to say that it doesn't matter what GPU reviewers use to compare AMD and Intel CPUs and it doesn't matter what CPU reviewers use to compare AMD and Nvidia GPUs either. It's all fair game.
Click to expand...

If I didn't have money to burn, which I don't, and I was building a gaming system today intending it to last for the next three, four or possibly even five years, then I'd invest in the Ryzen 5 1600 rather than the more expensive Core i7-7700K, especially if $500+ GPUs aren't in your future.
Click to expand...
With less powerful GPU's the FPS gaps disappear, and for the average budget gamer using a midgrade card you won't notice much difference between Ryzen and Intel depending on the titles.
We also saw how much more the higher 1440p resolution brings both Ryzen and Vega into play. Ryzen still did well at 1080p for the most part, though Vega certainly appears much more competitive at 1440p versus 1080p.
Click to expand...
Interesting with Vega that the 1600 did beat the 7700k in a few games at 1440P. Maybe with 7nm Vega and some availability we could see Ryzen doing very well in the future at higher resolutions.
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom