Soleil

Many thanks for your time !

Actually yes. Then if you do, I'll accept it.

That's neat. Please update your answer with this so I can accept it.

I guess that's the end word.

I'm well aware. I want to run Legacy with WSL2, not "Ubuntu20.04"

@Ramhound I still have "There is no distribution with the supplied name." and Legacy is still set to WSL1.

@Ramhound yes, but uname -r still gives "4.4.0-19041-Microsoft" (ie., WSL 1).

@Ramhound As mentioned, I'm not using the regular ubuntu distribution from the store, but the previous one (from before they appeared on the store), so it's not clear how to achieve what I want. I also updated the post.

@Scratte Got carnival too. I think it's about wearing and changing hat for x consecutive days. I wore the same for all SE sites, wore one everyday, but did not change everyday.

@DavidPostill what's the trigger for balalaika ?

@MattDMo "If you earned themm, they will be awarded to you again shortly. " chat.stackexchange.com/transcript/message/56467555#56467555

@YaakovEllis So, among those who had Disciple and Balalaïka, there are those who earned it, and those who did not, and after the update, only those who earned them will have them back?

@Catija is "wearing != having" based on some hat functions/definitions ? In everyday situation, wearing is having (on the head) ;)

@Catija yes few have still it meta.stackexchange.com/users/905312/cettt

@YaakovEllis Ellis, don't you unit test your hat functions?

Anyway. Thanks for your time.

BTW I'm not sure we can get RTX/Volta GPUs already on second hand market already.

Also probably the Formula.

I'm almost certain it will be 9900k + 1-2 2080ti, rather than a 9940x + 0-1 2080

You challenged me on the whole solution, and I had to explain the requirements. I thank you for that. This helps me to reconsider things.

And I really don't see myself transmitting 10-60GB for each workload (once every few minutes)

I'm also using the new gear 100+ h per week... So it will be cheaper at home.

The original question sounds quite simple in the end. And not that much important,

I got about 4-5k as budget for the update (new WS), so, no tesla.

Compute with AVX512 can be 4x faster than without, this depends on my code.

I have also a fair amount of compute on 1 core (the UI thread), hence the 9900k choice.

I own a 6700k on a matx, and I'm looking for something faster and more expandable.

The problem with the GPU is that I need mostly a lot of memory in 2 cards, and also the FP16, which is only with the RTX generation.

I'm having also loads of unit tests running as I code, so I need some really reactive CPU. I have a 6700k that is sluggish, and it can't even play a 4k H265 video. And I'll be processing loads of things like that.

Such as 4xTitan RTX etc

It's also for a sprint of few month, to do reasearch and MVPs. Later I might go with bigger things.

Indeed. And time (cuda, MKL code etc)

What would be the 'sunk cost fallacy' ?

So, for the same price I prefer a 9900k + 2080ti than a 9940x alone

indeed.

Yes ineed.

Yes, I was considering a lot the 9940x + x299 + 128GB instead of a 9900k + 64GB

I already got the server (Intel). I also have the Parallel Studio XE (~4k$.

I understand. I might really want to consider a skylake X due to the AVX512

It's not the point. The point is the MB.

MKL, opencv, Parallel studio XE. 370 (formula) vs 280 + 90 (hero + NIC)

I might add that that price is the same in both case !

Deep learning happens almost fully inside the GPU. Computer vision is more affected by the presence of AVX512, and number of cores. Amd has bandwidth problems at the cache level, and has poor driver/library support. Anyway, here, it's only about the motherboard + NIC.

@JMY1000 Budget is fixed since the price(Formula) = price(Hero+NIC 10Gbit). In the first case I got 5Gbit and 32 lines, in the second 10Gbit and 28 lines. That sums up the situation.

I updated, It's for a 9900k based WS, so no Zen2 for this guy. I'm using loads of code optimized for Skylake and the Intel compiler. Indeed, I wish I could go for a 9940x for instance.

Budget is relatively limited (I wish I could go for a SkylakeX or even few DGX2). It's a 1 WS and 1 server (headless). The WS will be dedicated to deep learning, computer vision and software development. The server will hold a hyperconvergent platform (many VMs) based on Windows server 2019 datacenter; and it will need to stay connected 24/7. The WS should be able to be rebooted or crashed.

Good luck

You should start here

Indeed