Ryzen review thread

Printable View

02-03-2017, 01:33 PM
CAT-THE-FIFTH

Ryzen review thread

R7 1800X

https://www.chiphell.com/article-17555-3.html (Chinese)
http://3c.3dmgame.com/show-52-5363-1-all.html (Chinese)

https://www.guru3d.com/articles-page...-review,1.html
http://hexus.net/tech/reviews/cpu/10...nm-zen/?page=9
http://www.hardwarecanucks.com/forum...ce-review.html
http://www.pcgameshardware.de/Ryzen-...eview-1222033/
http://www.eteknix.com/amd-ryzen-7-1...cessor-review/
https://arstechnica.co.uk/gadgets/20...-ryzen-review/
https://www.hardwareluxx.de/index.ph...x-im-test.html
http://www.kitguru.net/components/cp...0x-cpu-review/
http://www.pcworld.com/article/31757...amd-power.html
https://www.bit-tech.net/hardware/20...tform-review/7
http://www.anandtech.com/show/11170/...00x-and-1700/3
http://techreport.com/review/31366/a...-cpus-reviewed
https://www.pcper.com/reviews/Proces...ew-Now-and-Zen
http://www.hardwarecanucks.com/forum...ce-review.html
http://www.tweaktown.com/reviews/807...ady/index.html
http://www.gamersnexus.net/hwreviews...fps-benchmarks
http://www.tomshardware.co.uk/amd-ry...iew-33811.html
http://pclab.pl/art72996.html
http://www.pcgameshardware.de/Ryzen-...eview-1222033/

https://www.youtube.com/watch?v=u3eCASYfg-Y

R7 1800X,R7 1700X

http://www.sweclockers.com/test/2342...0x-och-7-1700x
https://nl.hardware.info/reviews/722...tie-voor-intel

R7 1700X

http://www.hardocp.com/article/2017/...x_cpu_review/1
http://www.guru3d.com/articles-pages...-review,1.html
http://hexus.net/tech/reviews/cpu/10...700x-14nm-zen/

R7 1800X,R7 1700X and R7 1700

http://www.overclockersclub.com/revi...00x_1700/9.htm
http://www.legitreviews.com/amd-ryze...-review_191753
https://www.computerbase.de/2017-03/...00x-1700-test/

R7 1700 overclocking review

http://www.xtremesystems.org/forums/...rn-of-the-Jedi
http://www.legitreviews.com/amd-ryze...ocessor_192191

R7 1800X Linux review

http://www.phoronix.com/vr.php?view=24219

R7 1700 server review

https://www.servethehome.com/amd-ryz...ux-benchmarks/

In-depth Ryzen 7 technical analysis in non-gaming scenarios

https://forums.anandtech.com/threads...nical.2500572/

IMPORTANT INFO.

SMT was not deactivated in many reviews leading to regressions in gaming performance:
http://i.imgur.com/UFaWvLe.jpg

http://i.imgur.com/UFaWvLe.jpg

The regressions are reasonably noticeable and many review sites are not aware of this.

Ryzen was also not launched with Windows drivers:

http://www.legitreviews.com/amd-ryze...view_191753/15

Quote:

Originally Posted by Legitreviews guy

You hit the nail on the head on all your comments. I really wish the Windows drivers were ready before we were given the parts to review. Instead we got a statement 24 hours before launch from AMD saying that they'll be coming in 30 days if all goes well. Game optimizations will be hit or miss when they come, but they appear to be coming. That takes time though and we'll see what happens.

Quote:

I put it in the conclusion on the last page. The quote came direct from AMD's John Taylor. There was talk of it coming with Ryzen 5 and then they said 'in the next month' in an official statement that was e-mailed out last night.

So there might be room for some better performance figures as time progresses.

Update!!

The SMT issue is down to Windows 10. It looks like The Stilt has done even more testing:

https://forums.anandtech.com/threads...#post-38775732

Quote:

I did some 3D testing and eventhou there is not nearly enough data to confirm it, I'd say the SMT regression is infact a Windows 10 related issue.
In 3D testing I did recently on Windows 10, the title which illustrated the biggest SMT regression was Total War: Warhammer.

All of these were recorded at 3.5GHz, 2133MHz MEMCLK with R9 Nano:

Windows 10 - 1080 Ultra DX11:

8C/16T - 49.39fps (Min), 72.36fps (Avg)
8C/8T - 57.16fps (Min), 72.46fps (Avg)

Windows 7 - 1080 Ultra DX11:

8C/16T - 62.33fps (Min), 78.18fps (Avg)
8C/8T - 62.00fps (Min), 73.22fps (Avg)

At the moment this is just pure speculation as there were variables, which could not be isolated.
Windows 10 figures were recorded using PresentMon (OCAT), however with Windows 7 it was necessary to use Fraps.

http://www.pcgameshardware.de/Ryzen-...eview-1222033/

It seems the 4C versions which appear to use only one CCX might be fine.

It looks like Hardware.fr has done some more testing with SMT off.

http://www.hardware.fr/getgraphimg.php?id=451&n=1

But here is the big kicker!!

http://www.hardware.fr/getgraphimg.php?id=453&n=1

They tested in performance mode and found not all the regressions were actually only down to SMT! ;)

08/03/2017 Update

Another day another potential issue which might have hampered performance at launch:

http://hwbot.org/newsflash/4335_ryze...ect_benchmarks

Quote:

Originally Posted by der8auer

Just ban Gigabyte boards for using old AGESA and make mainboard tab mandatory for AM4 submissions. Not sure about MSI or ASRock but it's not possible to do this on ASUS.

I seriously think with any AMD launch you need +3 months added to the launch to actually get an accurate picture!! :p
02-03-2017, 02:10 PM
Jowsey

Re: Ryzen review thread

I've just clicked it and I have my first question.

...Fish?
02-03-2017, 02:18 PM
CAT-THE-FIFTH

Re: Ryzen review thread

Quote:

Originally Posted by Jowsey

I've just clicked it and I have my first question.

...Fish?

https://www.youtube.com/watch?v=dkjbMoj0JY4#

:p
02-03-2017, 02:21 PM
scaryjim

Re: Ryzen review thread

My Q: are chiphell going to claim they simply miscalculated the NDA expiry time, or are they going to front up to the fact that they broke embargo by several hours...?
02-03-2017, 02:25 PM
CAT-THE-FIFTH

Re: Ryzen review thread

Quote:

Originally Posted by scaryjim

My Q: are chiphell going to claim they simply miscalculated the NDA expiry time, or are they going to front up to the fact that they broke embargo by several hours...?

Boost is probably not fully working on their sample or the CPU might be TDP throttling,as people who bought the R7 1700 are having better CB R15 scores when overclocked:

https://uploads.disquscdn.com/images...b03b22ef9a.jpg

https://uploads.disquscdn.com/images...b03b22ef9a.jpg

Their score is 148 points for the ST benchmark.
02-03-2017, 02:50 PM
CAT-THE-FIFTH

Re: Ryzen review thread

The R7 1700 looks a very decent overclocker:

https://forums.overclockers.co.uk/th...king.18771151/
02-03-2017, 03:21 PM
kalniel

Re: Ryzen review thread

http://www.anandtech.com/show/11170/...1700x-and-1700
http://hexus.net/tech/reviews/cpu/10...800x-14nm-zen/
02-03-2017, 05:58 PM
CAT-THE-FIFTH

Re: Ryzen review thread

OP updated.
02-03-2017, 06:18 PM
CAT-THE-FIFTH

Re: Ryzen review thread

OP updated.
03-03-2017, 12:32 AM
CAT-THE-FIFTH

Re: Ryzen review thread

OP updated with Linux review and overclocking review.
03-03-2017, 03:33 AM
CAT-THE-FIFTH

Re: Ryzen review thread

Added a server review.
03-03-2017, 09:51 AM
Phage

Re: Ryzen review thread

1) Cat - 2:30 am ? Are you in Australia ? If not,
https://img.clipartfest.com/2db5fed5...t_320-353.jpeg

2) That 1700 OCUK article looks very promising. I'm going to give it a month or so and see where the cards fall then. I suspect that Zen may well be the next system
03-03-2017, 11:05 AM
jim

Re: Ryzen review thread

I know I'm being *slightly* lazy here by not reading all of the above, but has anyone either done or said they will do a DDR4 speed benchmark with Ryzen?
03-03-2017, 11:25 AM
kalniel

Re: Ryzen review thread

Quote:

Originally Posted by jim

I know I'm being *slightly* lazy here by not reading all of the above, but has anyone either done or said they will do a DDR4 speed benchmark with Ryzen?

Anand say they will.

Quote:

Originally Posted by anandtech

For Ryzen, AMD is taking the latter approach. What we have is DDR4-2666 for the simplest configuration – one module per channel of single rank UDIMMs. This moves through to DDR4-1866 for the most strenuous configuration at two modules per channel with dual-rank UDIMMs. For our testing, we were running the memory at DDR4-2400, for lack of a fixed option, however we will have memory scaling numbers in due course
03-03-2017, 01:06 PM
kompukare

Re: Ryzen review thread

This youtube reviewer only had BF1 benched but ran it at max settings with Shadowplay on:
https://www.youtube.com/watch?v=sciuiEcrnzg
The minimum was better on the Ryzen than on the Kabylake, plus the Ryzen had over twice the mins of the FX-8300:
https://i.imgur.com/q8ibjVj.png
Despite the slow RAM they used.
Sort of what you'd expect in a way: more real cores should be great for streaming and background tasks. Problem is that's only one game and most reviewers run on a clean min install with nothing in the background whereas most people have stuff going on in the background (I know some people have kill-all-tasks type batch and change their computer's energy profile each time they game but that's not that common).
03-03-2017, 01:27 PM
CAT-THE-FIFTH

Re: Ryzen review thread

Anyway,since Ryzen is not only about gaming this review by The Stilt on the none-gaming abilities of Ryzen is probably one of the most in-depth released so far and covers ground many other reviews have not:

https://forums.anandtech.com/threads...nical.2500572/
04-03-2017, 12:14 AM
Gunbuster

Re: Ryzen review thread

Interestingly Windows Power Profile seems to matter, as seen in AMD Ryzen 7 1800X & 1700X Review: Live Up to The Hype? by Hardware Unboxed
05-03-2017, 01:45 AM
CAT-THE-FIFTH

Re: Ryzen review thread

OP updated!!
05-03-2017, 01:34 PM
scaryjim

Re: Ryzen review thread

Quote:

Originally Posted by Gunbuster

Interestingly Windows Power Profile seems to matter ...

WCCFTech have a great follow up article with some information from AMD about some of the problems (originally shared by anselhelm):

http://wccftech.com/amd-ryzen-launch...-amd-response/

Apparently in balanced power mode the clock/voltage updates only occur once every 30ms, while on the high performance mode it updates every 1ms. The new windows driver should fix that.

Other issues: the Windows scheduler apparently has issues with Ryzen's cache size and distinguishing between true cores and virtual threads, so will often push threads onto the slower virtual threads when true cores are available; also High Precision Event Timers can be detrimental to Ryzen performance (perhaps they can also be disbled by the driver?). So known problems that can be fixed by drivers and scheduler updates.
05-03-2017, 01:45 PM
CAT-THE-FIFTH

Re: Ryzen review thread

Quote:

Originally Posted by scaryjim

WCCFTech have a great follow up article with some information from AMD about some of the problems (originally shared by anselhelm):

http://wccftech.com/amd-ryzen-launch...-amd-response/

Apparently in balanced power mode the clock/voltage updates only occur once every 30ms, while on the high performance mode it updates every 1ms. The new windows driver should fix that.

Other issues: the Windows scheduler apparently has issues with Ryzen's cache size and distinguishing between true cores and virtual threads, so will often push threads onto the slower virtual threads when true cores are available; also High Precision Event Timers can be detrimental to Ryzen performance (perhaps they can also be disbled by the driver?). So known problems that can be fixed by drivers and scheduler updates.

I mentioned this in the Hexus R7 1800X review thread already:
http://forums.hexus.net/hexus-review...ml#post3776632

I also mentioned the issue about cache and thread priority here:

http://forums.hexus.net/hexus-review...ml#post3777796

Look at the top two posts.
08-03-2017, 06:34 PM
CAT-THE-FIFTH

Re: Ryzen review thread

Another day another potential issue which might have hampered performance at launch:

http://hwbot.org/newsflash/4335_ryze...ect_benchmarks

Quote:

Originally Posted by der8auer

Just ban Gigabyte boards for using old AGESA and make mainboard tab mandatory for AM4 submissions. Not sure about MSI or ASRock but it's not possible to do this on ASUS.

I seriously think with any AMD launch you need +3 months added to the launch to actually get an accurate picture!! :p