Looking at the AMD Threadripper units - General questions

[liberty610]

Hey guys,

I am currently running a fairly new system (built it last November). I have the Intel 6800k CPU with a Corsair Corsair Hydro Series H115i cooler. Here is my current entire setup via PC Parts Picker:

https://pcpartpicker.com/user/liberty610/saved/#view=G48RsY

I have started reading up on the AMD Threadripper, and I am VERY interested in the AMD RYZEN Threadripper 1950X 1(6-Core / 32 Threads). I have a beautiful board picked out for it as well.

https://www.newegg.com/Product/Product.aspx?Item=N82E16819113447&cm_re=AMD_Ryzen-_-19-113-447-_-Product

https://www.newegg.com/Product/Product.aspx?Item=N82E16813145030&cm_re=x399-Gaming_7-_-13-145-030-_-Product

My question is, what is the cooler ordeal for these bigger Threadripper CPUs? I just bought that Corasir luquid cooler, and I really don't wanna shelf out for another cooler. But I know the threadripper CPUs are way bigger. Not sure if Corsair is releasing an upgrade for this cooler for it or not....

Any thoughts?

 

[liberty610]

Yea, apparently the Gigabyte app doesn't pull the latest version of the apps that control things like the RGB stuff from their servers.... which I thought was odd, because the whole point of the app was to get you the latest versions of everything so you didn't have to keep going to the website and scrolling through everything to get what you are looking for. So I went to the site, got the latest everything i wanted to install, and we have no more crash from that issue.

And yea, I have plenty of ram sticks that are laying around. Mostly never going to get used again because, like you said, they still work, but are DDr2, DDR3 that i could sell, ect.

When you run memtest, is there are setting where you can test each stick by itself? Or do you have to pull them out of the slots and do it? I'm assuming you can tell it to check each stikc individually and then do a full torture test from within the interface?

 

[liberty610]

Okay guys, got another blue screen, this time while trying some rendering In Vegas Pro 14.

The stop error code I got was system threat exception not handled.

Couple things I've tried so far, I have updated my video drivers because they were not the latest versions like I thought they were. Gigabyte tech-support is telling me to try to take off XMP file on the ram and let the CPU process the memory instead of trying to overclock the memory with xmp. I have not tried that yet.

The other thing that I have don is, I change the amount of dynamic RAM preview that was set in Vegas's properties. I have 32 GB of RAM available, and I think the dynamic preview setting was 28 GB. I am not sure why it was up that high, but I pulled it down to only using half of my 32. So it is now set to 16 GB.

The odd thing about this crash, was after the blue screen went off and the computer rebooted, my motherboard beep three times and then powered itself off again(I have mobo speaker installed). It turned itself back on beep three times and then turn itself back off again. And then booted up, and posted just fine. In the heat of the moment, I could not really recall if it was three short beeps or three long peeps. But when rendering about 1/2 hour long video of full HD footage, it only got about 15 minutes into it then crashed.

So far I have changed the dynamic RAM preview allowance in Vegas down to 16 GB instead of 28. I have also updated my graphics drivers to the latest version. I did not turn off XMP profile yet. I have started rendering the video again, and so far it has gone to 53% complete and is been running for about 27 minutes.

The entire time that I was rendering, I had NZXT cam software opened up as well as hardware monitor opened up.I did not see a CPU temperature above 55°C. I believe the hovered around 53 or 54.So I cannot imagine that this is an overheating issue of the CPU.

Is it possible that allowing Vegas to access that much of my RAM would cause this kind of crash? I again have lowered the amount of RAM that Vegas can access for the dynamical preview down to 16 GB.

 

[Solarion]

Okay guys, got another blue screen, this time while trying some rendering In Vegas Pro 14.

The stop error code I got was system threat exception not handled.

Couple things I've tried so far, I have updated my video drivers because they were not the latest versions like I thought they were. Gigabyte tech-support is telling me to try to take off XMP file on the ram and let the CPU process the memory instead of trying to overclock the memory with xmp. I have not tried that yet.

The other thing that I have don is, I change the amount of dynamic RAM preview that was set in Vegas's properties. I have 32 GB of RAM available, and I think the dynamic preview setting was 28 GB. I am not sure why it was up that high, but I pulled it down to only using half of my 32. So it is now set to 16 GB.

The odd thing about this crash, was after the blue screen went off and the computer rebooted, my motherboard beep three times and then powered itself off again(I have mobo speaker installed). It turned itself back on beep three times and then turn itself back off again. And then booted up, and posted just fine. In the heat of the moment, I could not really recall if it was three short beeps or three long peeps. But when rendering about 1/2 hour long video of full HD footage, it only got about 15 minutes into it then crashed.

So far I have changed the dynamic RAM preview allowance in Vegas down to 16 GB instead of 28. I have also updated my graphics drivers to the latest version. I did not turn off XMP profile yet. I have started rendering the video again, and so far it has gone to 53% complete and is been running for about 27 minutes.

The entire time that I was rendering, I had NZXT cam software opened up as well as hardware monitor opened up.I did not see a CPU temperature above 55°C. I believe the hovered around 53 or 54.So I cannot imagine that this is an overheating issue of the CPU.

Is it possible that allowing Vegas to access that much of my RAM would cause this kind of crash? I again have lowered the amount of RAM that Vegas can access for the dynamical preview down to 16 GB.

Sounds like a memory fault(warm boot issue), but without more information it's hard to say. Have you looked up post codes for your particular board? I can't imagine the memory setting in vegas bringing down the whole system, it should have just resulted in swapping if there wasn't enough physical ram available.

This kind of thing is also why I'll likely be spending days hunting down the highest stable memory settings before doing much else with the platform. Memory problems can look like a plethora of issues, so you end up wasting a lot of time tracking down problems elsewhere.

I test individual memory sticks only when they're right out of the package. Once they've all passed and all are installed, they generally stay there, unless there's a problem.

BTW, now that my own board is one day out, I may not be on an MSI board for very long. Seems they "forgot" to enable a switch for SVM(virtualization) in the BIOS for the x399. Apparently it's being reported as present by windoze and linux, but disabled in BIOS with no switch provided to enable it. Oops. Doubtless they'll get it fixed eventually, but that's like half of my workload that won't run until it gets fixed. Quite the gaff on their part. Unfortunately the issue didn't show up on MSI forums until a day after my board had shipped.

 

[RobCrezz]

Yeah I agree, try it with XMP off, get it stable first.

 

[liberty610]

Thanks guys.for the replies. I was able to render the video down last night after it rebooted, and I ran handbrake on the rendered video afterwards, as well as did a lot of file moving off my production gadgets SD card. Didn't have any issues at that point.

I left the computer up all night just to see if there would be anything odd this morning. No issues when I got up. Went to work, was using the NZXT cam software on my phone to monitor the temps on my rig. About 6am, cam app on my.phone said pc was off. So Idk if it crashed while I was at work, or if there are issues with Cam's servers or something. Heading home soon to see if task manager can tell me the up time since my last boot.

And Solarion, that sucks abhor your board. But again, doesn't surprise me. MSI is only in my good graces with their graphics cards. Their boards or nothing but trouble.

 

[Solarion]

And Solarion, that sucks abhor your board. But again, doesn't surprise me. MSI is only in my good graces with their graphics cards. Their boards or nothing but trouble.

Thanks. Such a glaring oversight, really shocked that they missed it. On the plus side, a minimal amount of griping on their support forums resulted in a fix for the missing setting within a day...so that was great to see.

Now my only issue with the MSI is that they haven't added support for my particular NVME drive. The MyDigitalSSD BPX. To be fair it is not on MSI's compatibility list, so not really an oversight, but it's my favorite low cost drive, and it's not recognized on the x399 Carbon. Now I either have to pester MSI till they add support or swap it out for the Sammy 960Pro which, while it is a good drive, costs around 25% more/GB. I suppose if I'm going through the hassle of returning an m.2 drive, I may just use it as an excuse to go to a larger 1TB drive.

 

[RobCrezz]

Does that drive not work on it then? Just because they don't list it as supported doesn't mean it wont work fine.

 

[Solarion]

Yeah, that's why I gave it a shot...despite it not appearing on the QVL, but unfortunately it doesn't show up in BIOS on the MSI x399. I've now tried 4 different BIOS revisions from MSI with no change on this front. I'm going to pester MSI a bit to see if I can get support for this ridiculously awesome price/performance drive, but I may end up returning it.

According to the drive manufacturer it is compatible with the the MSI z270 SLI Plus, and I have one of those around, so I'm going to drop it in there to ensure it's not a problem with the drive itself.

 

[RobCrezz]

Ah that sucks!

 

[Solarion]

Doesn't work in MSI z270 SLI Plus either. At this point I'm assuming the drive is faulty and returning it for exchange. The real question is...exchange for same or for something that's on the mobo manufacturer's QVL. Wish the Sammy 960 EVO didn't have such a horrible TBW and/or the 960 Pro wasn't so overpriced. The BPX has a TBW rating that's 3.5x that of the EVO and costs 32% less than the 960 Pro/GB. Decisions decisions...