Ryzen 3950X, 3960X, 3970X Availability

So with all these CPUs popping up in stock (and disappearing almost as fast) it looks far better than the Intel offering...has anyone managed to find a i9 10980XE in stock, anywhere?

The only way the Intel launch could have gone worse is if there were woodscrews securing the IHS
 
My mobo showed up today, but my CPU is MIA. Hasn't moved on Fedex tracking in 24 hours when it was scheduled to get here today...oh well. I was hoping to at least test the mobo/cpu on a bench real quick before I go out of town for Christmas tomorrow but it apparently wasn't meant to be. I'll be back with a build on the 29th or so I guess.
 
Have not been able to get my hands on the big boys.This will do in interim.

DAN D.Hows the msi unify review coming?Makes asus board look like a bitch.
 

Attachments

  • 61145-.PNG
    61145-.PNG
    223.9 KB · Views: 0
Wow Derbauer temps are awful af. Even stock they're a joke.

I can max my chip out at full p95 avx full.load small fft and it pass 67c.

AMD did a knockout job at soldering the IHS. Now as far as the heat killer block, throw that shit in the trash and get a real block like XSPC.
 
Wow Derbauer temps are awful af. Even stock they're a joke.

I can max my chip out at full p95 avx full.load small fft and it pass 67c.

AMD did a knockout job at soldering the IHS. Now as far as the heat killer block, throw that shit in the trash and get a real block like XSPC.

I don't know what you are talking about. The Heatkiller IV is supposed to be one of, if not the best block money can buy.

XSPC on the other hand - from personal experience - I don't trust any of their shit to not explode and french everything in coolant. I almost consider them worse than no-brand chinese parts. They would be one of my last choices.
 
Optimums new king on the block.

https://optimuspc.com/

First I've ever heard of them.

As a rule of thumb I only trust established brands for water-cooling.

Swiftech, Alphacool, Watercool.de, even EK (sometimes, their first rev Threadripper block was a disaster) Maybe if they have a good track record in a decade or so when I need a new block I'll consider them :p
 
Last edited:
First I've ever heard of them.

As a rule of thumb I only trust established brands for water-cooling.

Swiftech, Alphacool, Watercool.de, even EK (sometimes, their first rev Threadripper block was a diaaster) Maybe if they have a good track record in a decade or so when I need a new block I'll consider them :p

Look where that got the dinosaur.Adapt or parish.
 
“All us overclocking geeks have always known that surface area is king when it comes to cooling and Optimus has surely proven that.”

HardOCP Review

link is no longer but wayback machine works.. looks like Kyle reviewed it back in 2017' ?
 
“All us overclocking geeks have always known that surface area is king when it comes to cooling and Optimus has surely proven that.”

HardOCP Review

link is no longer but wayback machine works.. looks like Kyle reviewed it back in 2017' ?

Wow. How did I miss that?

That's a couple of years of history, and a trusted reviewer.

I may be warming up to it.

You have to be cautious in the YouTube era. Can't trust any of those fuckwit excuses for reviewers on YouTube. You never know who is a paid "influencer".
 
I don't know what you are talking about. The Heatkiller IV is supposed to be one of, if not the best block money can buy.

XSPC on the other hand - from personal experience - I don't trust any of their shit to not explode and french everything in coolant. I almost consider them worse than no-brand chinese parts. They would be one of my last choices.

Did you have one blow up?
 

Attachments

  • upload_2019-12-17_20-44-40.png
    upload_2019-12-17_20-44-40.png
    12.2 KB · Views: 0
Last edited:
Did you have one blow up?

Der(I dont feel like spelling his gamer name) was reporting stock temps @ over 80c sometimes pushing 90c So... come again how is heatkiller that good?

The 3960x, of which he used, is factory rated from AMD @ 68c --- mine stays under that with my shitty leaky XSPC raystorm. Ive posted several screenshots of temps, loads, etc.... not going to do it again.

View attachment 208209

Kyle Bennet said that he loved the Raystorm. Can't find the review of it anymore on OCP because they changed the website to the FPSreview.

I could go on and on but this takes the thread too far off topic from the availability of Ryzen as the actual topic.

All i'm going to say is that this is a Ford v. Dodge v. Chevy conversation that no one wins or can win. If you want to run that ugly af heatkiller that got Der8 those awful temps pre and post delid then be my guest. I am happy with my 4 different XSPC blocks I own that not a single one has leaked, dripped, cracked, or had some weird bi-metal ionization failed nickel plating corrosion failure that EK and others have dealt with.

I dont buy the internet hype that something sucks until I try it myself unless the claims are ginormously substantial in size. Maybe you just bought the hype. I dont know. I make no claims or blames brother!

XSPC has been out long before the Heatkiller VI etc was even on the drawing board.

I have had some pretty nasty leaks with XSPC parts. Not blocks specifically, but they gave me a really bad taste in my mouth. Nothing else I've ever used has leaked on me.

There is no doubt there is something wrong with the der dudes setup, but I don't think the water block is it. As I recall the Heatkiller IV pro did best of all the blocks Kyle tested in his Threadripper waterblock reviews. Sucks that these are all down now. The first page of the reviews seems to be all that has been archived on the Internet Archive :(
 
  • Like
Reactions: N4CR
like this
I have had some pretty nasty leaks with XSPC parts. Not blocks specifically, but they gave me a really bad taste in my mouth. Nothing else I've ever used has leaked on me.

There is no doubt there is something wrong with the der dudes setup, but I don't think the water block is it. As I recall the Heatkiller IV pro did best of all the blocks Kyle tested in his Threadripper waterblock reviews. Sucks that these are all down now. The first page of the reviews seems to be all that has been archived on the Internet Archive :(

Yeah I have always heard Heatkiller are good but never had the oppurtunity to try one.

And you are right! There was something awful with Der's setup. No way temps should have been that high. I will take back my direct attack on Heatkiller because I am not qualified by experience directly with one to make such a claim.

I have tried EK, Byski, XSPC, Corsair, and a few others, but never Heatkiller. All of them perform quite well except EK was awful. I would not recommend their threadripper block. I am sure it has improved.

I still think the heatkiller is one FUGLY block hahaha so you go right ahead Zara :p
 
Yeah I have always heard Heatkiller are good but never had the oppurtunity to try one.

Yeah, this will be my first as well!

And you are right! There was something awful with Der's setup. No way temps should have been that high. I will take back my direct attack on Heatkiller because I am not qualified by experience directly with one to make such a claim.

I suspect something else is going on. Either he has poor block mount, poor flow or his fans are off or not working right. Or maybe he is doing something stupid and giving it way too much voltage. Or maybe he got a bad block, CPU or motherboard.

I know he is highly regarded these days (I still don't trust anyone on YouTube, but that's just me) but no one is perfect.

I have tried EK, Byski, XSPC, Corsair, and a few others, but never Heatkiller. All of them perform quite well except EK was awful. I would not recommend their threadripper block. I am sure it has improved.

I have generally had good experiences with EK. At first I wanted to avoid them after the whole nickel corrosion issue, but when I first started out they made the only fullcover block I could find for my GPU, and once I decided I'd give the GPU block a try, I figured, in for a penny, in for a pound, so I got a Supremacy EVO block as well, as it was the top reviewing block in the reviews I could find. They have both been good to me. especially the GPU block. It has been absolutely amazing. When you have a Pascal Titan X heavily overclocked at full load sitting at 24C with full fans (37C with normal fans, with loop temp at 33C-34C) I can't imagine anything much better than that.

The Supremacy EVO all copper CPU block I have been less impressed by. My CPU temps stayed about the same at load with this block as they did with the Corsair H110i GTX AIO. I was expecting at least a little bit of an improvement.

I still think the heatkiller is one FUGLY block hahaha so you go right ahead Zara :p

I agree in regards to the plexi version. It really is quite ugly.

1015091-1_600x600.jpg



I quite like the solid metal ones, both the all copper and the nickel plated.

upload_2019-12-17_22-7-19.jpeg
upload_2019-12-17_22-7-40.jpeg


But to be honest, I don't build PC's for the aesthetics. I'm looking for the best possible performance at the lowest possible noise levels.

I went with the nickel plated version btw. Less tarnish to deal with. I had to order it from Germany as everyone was out of stock here in the U.S. And damn this block is surprisingly heavy, weighing in at 2.25lb
 
Zarathustra[H]

That copper one is nice looking! But if you are going Gigabyte mobo then go with Nickel because it will match a lot of the shiny chrome that Gigabyte loves to use.

@ General

About that Optimus Threadripper Block. Wow $199.00 ??

That thing must rock if they are willing to charge that much for it.

Well have to see what the reviews show if there are any.
 
We need HARDOCP back for good water block testing and reporting. No one did it better than Kyle.
 
  • Like
Reactions: N4CR
like this
Congratulations, Zara.

Anyone else surprised that huge box doesn't actually say what the specific item is? I can't tell if you got a 3960x or a 3970x. In any case, cool stuff, pretty jealous!
 
Congratulations, Zara.

Anyone else surprised that huge box doesn't actually say what the specific item is? I can't tell if you got a 3960x or a 3970x. In any case, cool stuff, pretty jealous!

Yeah, they put a window on the box. You can read the model on the top of the heat spreader (though it was too small in my picture)

I got the 3960x. I was more interested in higher clocks than I was in cores ad nauseum.
 
Wow Derbauer temps are awful af. Even stock they're a joke.

I can max my chip out at full p95 avx full.load small fft and it pass 67c.

AMD did a knockout job at soldering the IHS. Now as far as the heat killer block, throw that shit in the trash and get a real block like XSPC.

I just took a look at his video (I usually avoid YouTube like the plague, but I was curious)

Looks like his baseline temps in the 80's were run at an all core overclock.

I wonder if that explains his temps.

Even so, he did his baseline testing with Conebench which is pretty poor practice. That does not give a real thermal load to the system, so his temps - if anything - should have been lower than expected, not higher.
 
And since I don't want to add another post to show the F'd up TT waterblock innards. This is what all 4 corners of the block look like.

ThermalTake. Quality - Da'Fuk is that?! Who even?!

View attachment 208039

It's like they just machined it and hammered it into a tapered housing. What a shitty job.
P.s. QVL isn't a major issue with Zen2.. even Zen+ you would be pretty safe with most kits out there. But yes you want to run a decent amount of ram so better safe than sorry..
 
Wow Derbauer temps are awful af. Even stock they're a joke.

I can max my chip out at full p95 avx full.load small fft and it pass 67c.

AMD did a knockout job at soldering the IHS. Now as far as the heat killer block, throw that shit in the trash and get a real block like XSPC.

Nah, they are about right. Look at his settings, 1.4v and 4300 all core. When you move to an all core and add volts, these chips can become heat monsters. Btw, small is not = blend. 67c in p95 is not a real hardcore load.
 
Nah, they are about right. Look at his settings, 1.4v and 4300 all core. When you move to an all core and add volts, these chips can become heat monsters. Btw, small is not = blend. 67c in p95 is not a real hardcore load.


As I understand it Small FFT's keeps everything in CPU caches and thus hits the CPU HARD as it spends very little time in wait states. Because of this expect small fft to produce the most heat and will tease out any instability due soleley to the CPU.

It can be legitimately argued this load is a bit unrealistic as no matter what you do in the real world, at some point more data needs to be pulled in from relatively slow RAM or very alow disk.

I usually use blend. It is - IMHO - a more realistic high load scenario where the CPU enters wait states as it awaits more data from ram/disk. It puts a much higher load on the RAM than small fft does, so it should in theory be much better at catching any RAM instability.
 
Nah, they are about right. Look at his settings, 1.4v and 4300 all core. When you move to an all core and add volts, these chips can become heat monsters. Btw, small is not = blend. 67c in p95 is not a real hardcore load.

He made no mention of an all core

He needs to describe what hes doing better if he did

I don't use prime 95 for stability testing ever. I use h265 and other softwares.

It says right in p95 small fft = maximum stress and heat. Right there by the dev what it does. To assume another feature generates more heat means you are not trusting the team that made the software to do as claimed.
 
Last edited:
It's like they just machined it and hammered it into a tapered housing. What a shitty job.
P.s. QVL isn't a major issue with Zen2.. even Zen+ you would be pretty safe with most kits out there. But yes you want to run a decent amount of ram so better safe than sorry..

GN that did an alaysis on this - ram kits on the QVL often have subtimings set that are optimized and sometimes not exposed in the bios - meaning they will perform better because there are buried subtimings optimized for the kits. It was fairly amazing how much difference it made - 3200mhz kits that were on the QVL were performing on par with kits that were rated at 3866 but not on the list.
https://www.gamersnexus.net/guides/3508-ryzen-3000-memory-benchmark-best-ram-fclk-uclock-mclock


From the article:
"You'll notice going forward that the GeIL kit is always among the worst on the bench, but note one thing here: After refilming this twice for the Vengeance LPX kit retests, we also re-tested the GeIL kit for validation. It validated perfectly on the Gigabyte board, meaning that the GeIL kit remained in exactly the same spot in the charts. We then tested it on the MSI board, and found that MSI's board auto-configured better timings than Gigabyte for this GeIL memory to significantly improve performance. "

It was on the QVL for the MSI board but not the gigabyte board. I know a lot of boards for these have all of the subtimings exposed, but if you're more of a set docp and forget it, I'd stick to the QVL, especially for higher priced kits as otherwise you might not be getting what you paid for.
 
It says right in p95 small fft = maximum stress and heat. Right there by the dev what it does. To assume another feature generates more heat means you are not trusting the team that made the software to do as claimed.

Small is not the highest stress. Look at the options, blend = everything and the kitchen sink. Small is high heat and keeps load inside cpu, doesn't even stress memory/imc. Small is not a realistic load, can pass small and crash elsewhere.
 
Small is not the highest stress. Look at the options, blend = everything and the kitchen sink. Small is high heat and keeps load inside cpu, doesn't even stress memory/imc. Small is not a realistic load, can pass small and crash elsewhere.

Really depends on what you consider "highest stress". It puts much higher stress on the CPU itself than the other options, because the CPU spends less time in wait states waiting for the RAM or Disk.

It isolates that stress to just the CPU though so you can miss issues with other sibsystems, like RAM.
 
Back
Top