Threadripper 2 review thread

[CAT-THE-FIFTH]

2990WX

https://hexus.net/tech/reviews/cpu/1...ripper-2990wx/
https://www.guru3d.com/articles-page...-review,1.html
https://www.bit-tech.net/reviews/tec...950x-review/1/

2950X

https://www.techpowerup.com/reviews/...dripper_2950X/
https://www.guru3d.com/articles-page...-review,1.html
https://www.hardwarecanucks.com/foru...ce-review.html
https://hexus.net/tech/reviews/cpu/1...dripper-2950x/

2990WX and 2950X

https://techgage.com/article/amd-ryz...cessor-review/
https://techreport.com/review/33977/...x-cpu-reviewed
https://www.hardocp.com/article/2018...x_cpu_review/5
https://www.anandtech.com/show/13124...d-2950x-review
https://www.tomshardware.com/reviews...950x,5725.html
https://www.techspot.com/review/1678...-2990wx-2950x/
https://www.sweclockers.com/test/260...90wx-och-2950x
https://www.pcper.com/reviews/Proces...Moving-Forward
http://www.pcgameshardware.de/Ryzen-...-Test-1262448/
https://www.overclock3d.net/reviews/...990wx_review/5
https://www.computerbase.de/2018-08/...wx-2950x-test/


Linux reviews

https://www.phoronix.com/scan.php?pa...x-2990wx&num=1
https://www.phoronix.com/scan.php?pa...0x-linux&num=1
https://techgage.com/article/amd-ryz...ssor-review/6/

Windows vs Linux reviews

https://www.phoronix.com/scan.php?pa...-windows&num=1

Cooler reviews

https://www.phoronix.com/scan.php?pa...-cooling&num=1

Video reviews

Hardware Unboxed
https://www.youtube.com/watch?v=QI9sMfWmCsk

Linus Tech Tips
https://www.youtube.com/watch?v=KZKeDg4QVN4

Pauls Hardware
https://www.youtube.com/watch?v=YWYOqKhhX2A

Bitwit
https://www.youtube.com/watch?v=S02-DlKaIIU


Possible NV driver issue with 2990WX!



https://www.golem.de/news/32-kern-cp...08-136016.html
https://www.hardocp.com/news/2018/08...nvidia_driver/

[CAT-THE-FIFTH]

BTW,the AT reviewer forgot to remove the plastic off the cooler it appears for a number of tests.

[Biscuit]

BTW,the AT reviewer forgot to remove the plastic off the cooler it appears for a number of tests.

...wow

Apparently quite a few youtubers were not provided review samples for these chips, some of which have become quite salty about it.

https://www.reddit.com/r/Amd/comment...hreadripper_2/

[scaryjim]

Is it just me, or are Anandtech's performance figures all over the fricking place? And not just the AMD v Intel comparisons - within benchmarks some of the Intel-Intel and AMD-AMD comparisons make no sense whatsoever....

[Muaadib]

...wow

Apparently quite a few youtubers were not provided review samples for these chips, some of which have become quite salty about it.

https://www.reddit.com/r/Amd/comment...hreadripper_2/

Threatening to not use a CPU that you already have and changing to an inferior product just to protest not getting a free sample sounds extremely childish to me.

[virtuo]

.. youtubers ..

sounds extremely childish

Sounds about right

[Hoonigan]

Is it just me, or are Anandtech's performance figures all over the fricking place? And not just the AMD v Intel comparisons - within benchmarks some of the Intel-Intel and AMD-AMD comparisons make no sense whatsoever....

That's generally what happens when someone gets lazy and only runs tests once, instead of running 3 or 5 times and averaging each result.

That can be forgiven for tests like PCMark, which can take anywhere up to an hour to run, and they generally give a spot-on result first time, but things like Cinebench or POVRAY give no excuse.

[CAT-THE-FIFTH]

OP updated!!

This review seems to test video transcoding and rendering using multiple applications and even does Linux testing:

https://techgage.com/article/amd-ryz...cessor-review/

3. Encoding: Premiere Pro, Vegas, PhotoScan, Resolve & LameXP
4. Rendering: Arnold, Blender, KeyShot, Radeon ProRender & V-Ray Next
5. Rendering: Adobe Dimension, Cinebench, Cinema 4D, Corona, POV-Ray
6. Linux: Rendering, Compiling, Compression, & Cryptography

[watercooled]

Is it just me, or are Anandtech's performance figures all over the fricking place? And not just the AMD v Intel comparisons - within benchmarks some of the Intel-Intel and AMD-AMD comparisons make no sense whatsoever....

I've noticed that a few times - performance figures for apparently exactly the same test can vary wildly between tests. They're clearly from a different benchmark run but there's seldom an explanation why they've changed e.g. due to firmware updates, different test system, etc, but they're sometimes significant enough to completely change product rankings. Note the WD Black NVMe and Toshiba XG5:



So which is it? If the results are that unreliable then is it really a worthwhile benchmark?

The sustained results are also strange although they changed the methodology recently, presumably to control for difference in performance depending on drive fragmentation, but the results are still a bit strange, with the Black ending up higher even in the fragmented state vs the previous results!
https://www.anandtech.com/show/12543...d-ssd-review/7
https://www.anandtech.com/show/13080...ller-preview/6

Also, like you say some results are obviously outliers, and while they're possibly correct in some cases they should definitely get the attention of the reviewer. Again with the WD Black NVMe (just a drive I've been paying attention to) its low QD performance seems suspiciously low in these tests vs similar drives, and many reviewers just ignored it and moved on. However, Tweaktown spotted this and made a point of mentioning it:

The Phison E12 doesn't like our test and gives us a false reading at low queue depths. We had the same issue with the SanDisk Extreme Pro NVMe 3D SSD a few months ago. I wouldn't be surprised if the new E12 and SanDisk controllers don't share some background together. These are the first controllers we've had this issue with running low numbers at QD1 and QD2 since the Intel SSD 750 from three years ago. We know the low queue depth issue is a false reading because we can simply read data off the drive or run a simpler test to see higher performance. In HD Tune Pro the QD1 sequential read with the Phison E12 is 2,300 MB/s. We suspect the issue has something to do with the power management and the synthetic workload.

Read more: https://www.tweaktown.com/articles/8...en/index2.html

(The Sandisk is basically the same drive as the WD BTW)

Copying some files from the drive should have told them that something wasn't right about those synthetic results as the drive has no issues blowing right past that number for Windows file transfers.

[CAT-THE-FIFTH]

It looks like the NV driver might be the reason gaming performance seems a bit meh with the 2990WX:

https://www.golem.de/news/32-kern-cp...08-136016.html
https://www.hardocp.com/news/2018/08...nvidia_driver/



Loads of other games are also affected too it appears.

[watercooled]

Well spotted. You would think more reviewers would pick up on this - it's not like it's the first time this sort of thing has happened with drivers and a new release!

[Biscuit]

It looks like the NV driver might be the reason gaming performance seems a bit meh with the 2990WX:

https://www.golem.de/news/32-kern-cp...08-136016.html
https://www.hardocp.com/news/2018/08...nvidia_driver/



Loads of other games are also affected too it appears.

If I'm not mistaken, this was also an issue when comparing with Ryzen?

https://www.reddit.com/r/Amd/comment...e0&sh=482d7b0e

[CAT-THE-FIFTH]

If I'm not mistaken, this was also an issue when comparing with Ryzen?

https://www.reddit.com/r/Amd/comment...e0&sh=482d7b0e

Yep.

[Corky34]

Well spotted. You would think more reviewers would pick up on this - it's not like it's the first time this sort of thing has happened with drivers and a new release!

I'm more surprised that AMD didn't notice and put something in the reviewer guidance, that they didn't seems to point towards them either not testing with Nvidia hardware, it being a recent change in Nvidia drivers, or any number of things.

[CAT-THE-FIFTH]

I'm more surprised that AMD didn't notice and put something in the reviewer guidance, that they didn't seems to point towards them either not testing with Nvidia hardware, it being a recent change in Nvidia drivers, or any number of things.

Yeah,I made that comment to a mate too - AMD really needs to not miss on details like this,as my mate basically said initial reviews won't be updated with any newer information and he was correct.

[Biscuit]

You can imagine what reviewers would be saying if AMD specifically told them to use AMD cards with AMD CPUs in reviews though...