AMD Announces Ryzen Threadripper Pro 5000 WX-Series: Zen 3 For OEM Workstations

Can somebody describe the break in the trusted boot chain that they are trying to fix with the CPU fixed to the board?

What is the security problem they claim to fix?
This is AMD’s method for dealing with firmware persistent malware, root kits, and boot kits. It works in tandem with the AMD Secure Processor (ASP) which is a bit of silicon in their pro series chips that handles security and management. Similar to Intels Management Engine. It actually works pretty well in defending against Return-on-Programming attacks.
Since it works to encrypt all data moving across the system it also does well in defending or mitigating Spectre, Meltdown, and all those other side channel attacks that were going around.
 
Last edited:
Not sure if this old new or the good subject, but they have been shipped them for a while

 
I like how the benchmarks show an okayish gain, but real world use is night and day difference.
Not sure I see it, some real world (most-all?) use in that video do seem to match synthetic benchmark a lot

Chromium code compile 3.1% faster
blender render time 2.15% to 4.5% faster

Make you wonder if there is not an issue with their setup cooling wise or Windows instead of Unix-Linux being an issue .

Or maybe I am just jumping around the video without sound and misreading it.

Here:
https://www.tomshardware.com/news/amd-threadripper-pro-5995wx-5975wx-cpu-review

Lot of the time the 3xxx generation beat the 5xxxx, Classroom render faster on the 3990x than on the 5995wx or 5975wx.... that seem terrible no ?

Will see the Wendel reviews, maybe something is missing here.
 
Not sure I see it, some real world (most-all?) use in that video do seem to match synthetic benchmark a lot


A couple of minutes before are explanation of their video editing workflow and problems they were having. It went from unusable to smooth with the Threadripper Pro.
 
A couple of minutes before are explanation of their video editing workflow and problems they were having. It went from unusable to smooth with the Threadripper Pro.
Adobe pro puget benchmark had a giant difference has well (around 11:39) which again would show a good match between the benchmark and real world (I imagine puget benchmark are not synthetic but actual files doing actual work ?)
 
Adobe pro puget benchmark had a giant difference has well (around 11:39) which again would show a good match between the benchmark and real world (I imagine puget benchmark are not synthetic but actual files doing actual work ?)
This wasn't a "works similar, but takes less time" situation, this was binary. Went from not working to working. No benchmark they ran captured that
 
I think I lost the plot, the real work with a felt change (not sure why we would say not working to working, it was a current project they were able to work on with their current computer) had an almost 30% jump on the benchmark, maybe that what you meant by okish gain, but that seem very substantial.

Maybe we are saying the same thing here, but it seem to me were the benchmark had giant jump, relevant real world application got a jump, where the benchmark have terrible gain, real world also show terrible gain (sometime negative gain)
 
I think I lost the plot, the real work with a felt change (not sure why we would say not working to working, it was a current project they were able to work on with their current computer) had an almost 30% jump on the benchmark, maybe that what you meant by okish gain, but that seem very substantial.

Maybe we are saying the same thing here, but it seem to me were the benchmark had giant jump, relevant real world application got a jump, where the benchmark have terrible gain, real world also show terrible gain (sometime negative gain)
I don't believe you watched the video.

"The Threadripper 3000 series had no issues with 4K60 footage from the older GH5, but when we upgraded to the GH6 we wanted to start working with 5.7K60, and this proved troublesome. Of course, we could use proxies but they are a pain for editing, taking a lot more time, and overall just a more cumbersome method than we prefer.

We had been actively testing for alternatives, the Core i9-12900K was amazing at scrubbing through timelines for general editing, but as soon as we wanted to stabilize more than a dozen clips at once, the system became unusable and would often crash. Meaning we had to stabilize clips in batches, saving after each batch was complete before starting the next batch. It wasted a lot of time and the encoding process was also slower.

The first thing we did with the 5995WX was to open one of the current projects we were working on and start editing. The experience was totally different. At no point did the 5995WX struggle with the 5.7K footage like the 3990X did, even color-corrected footage with transitions played back smoothly.

In short, the 5995WX is a game changer for our workflow and as such it's difficult to put it into the context of percentage gains, it's not just faster, it's wildly more practical. You could say editing is now more playable, to use a gaming term.

We'd say we could easily justify spending $3,300 on the 5975WX, plus everything else you need to build the workstation PC, as it would pay itself off, both in terms of productivity and mental health.
"
 
Last edited:
Yes I saw that, again I am not sure I follow the plot, like I said maybe you did find a +30% result in benchmark Okish and not something that representative than a smoother playback, so I just miunderstood you, but I did not had the impression that it was impossible for a 3990x to do that workload, just that the playback was not smooth.

I felt your message implied that the benchmark were not representative of the real world performance change, which I am really unsure, real world compile, real world rendering, etc... seem in line with very okish benchmark, while Abode pro editing seem inline with a massive benchmark jump.

Which make me think that it is possible that they are not using them fully correctly, will see when the Wendel type test them under a Unix base system.
 
Can somebody describe the break in the trusted boot chain that they are trying to fix with the CPU fixed to the board?

What is the security problem they claim to fix?
I realize Im replying to an older post but the security idea is that no adversary can tamper with the BIOS and use the encryption keys inside the CPU to decrypt data, because the CPU would refuse to work with a modified non-vendor-signed BIOS.

Having the CPU just reset the encryption keys in such a case would be the way if it was *only* about security. Having the CPU refuse to work entirely seems clearly a move to stifle or eliminate secondary market activity (ebay/used/arbitrage/repurposing).
 
Last edited:
Back
Top