All things Nvidia GPU

Keith Myers
Keith Myers
Joined: 11 Feb 11
Posts: 4,802
Credit: 17,875,523,153
RAC: 2,709,578

If the app developers choose

If the app developers choose to code for the Blackwell architecture, we certainly could use the new gpus.

 

Ian&Steve C.
Ian&Steve C.
Joined: 19 Jan 20
Posts: 3,781
Credit: 36,888,081,919
RAC: 48,992,684

I'm sure they can use it on

I'm sure they can use it on their OpenCL variants as-is. it's just incredibly wasteful from a cost perspective. no one with $30-40k to spend on a single GPU is going to use it for this...

_________________________________________________________________________

Tom M
Tom M
Joined: 2 Feb 06
Posts: 5,913
Credit: 7,992,285,778
RAC: 6,270,900

Ian&Steve C. wrote: I'm sure

Ian&Steve C. wrote:

I'm sure they can use it on their OpenCL variants as-is. it's just incredibly wasteful from a cost perspective. no one with $30-40k to spend on a single GPU is going to use it for this...

Just imagine how many Rtx 3080 ti's or Titan V's you could buy for that.....   :)

A Proud member of the O.F.A.  (Old Farts Association).  Be well, do good work, and keep in touch.® (Garrison Keillor)

GWGeorge007
GWGeorge007
Joined: 8 Jan 18
Posts: 2,909
Credit: 4,791,793,662
RAC: 2,816,574

Tom M wrote: Ian&Steve C.

Tom M wrote:

Ian&Steve C. wrote:

I'm sure they can use it on their OpenCL variants as-is. it's just incredibly wasteful from a cost perspective. no one with $30-40k to spend on a single GPU is going to use it for this...

Just imagine how many Rtx 3080 ti's or Titan V's you could buy for that.....   :)

At an average price of $600 on EBAY for a used 3080Ti, 50+ units for $30,000.  Then, of course, you'll need more motherboards, PSUs, CPUs, RAM, SSDs, computer cases (maybe)...  I doubt that the next $10,000 will cover those items.  Need I go on?

George

Proud member of the Old Farts Association

Keith Myers
Keith Myers
Joined: 11 Feb 11
Posts: 4,802
Credit: 17,875,523,153
RAC: 2,709,578

Ian&Steve C. wrote: I'm sure

Ian&Steve C. wrote:

I'm sure they can use it on their OpenCL variants as-is. it's just incredibly wasteful from a cost perspective. no one with $30-40k to spend on a single GPU is going to use it for this...

I was referring to the eventual RTX5000 series consumer gpus that Nvidia will eventually produce.  I wouldn't hold my breath waiting for them though.

Nvidia's focus is on AI and the hardware for that.  That is tying up their contracted foundry utilization.  Will be a long while before they allocate foundry resources for cutdown Blackwell silicon for consumer gpus.

I'm guessing 3rd quarter 2025 before we see Blackwell RTX5090 gpus.

 

Tom M
Tom M
Joined: 2 Feb 06
Posts: 5,913
Credit: 7,992,285,778
RAC: 6,270,900

Are there any other strict

Are there any other strict two slot wide Nvidia GPU's?

Titan V

I am especially interested in rtx 3080 ti or faster GPU's.

So far the rtx 3080 ti's that are claiming 2 slots are still wider than a Titan V.

Tom M

A Proud member of the O.F.A.  (Old Farts Association).  Be well, do good work, and keep in touch.® (Garrison Keillor)

GWGeorge007
GWGeorge007
Joined: 8 Jan 18
Posts: 2,909
Credit: 4,791,793,662
RAC: 2,816,574

Tom M wrote: Are there any

Tom M wrote:

Are there any other strict two slot wide Nvidia GPU's?

Titan V

I am especially interested in rtx 3080 ti or faster GPU's.

So far the rtx 3080 ti's that are claiming 2 slots are still wider than a Titan V.

Tom M

The Hybrid GPUs are 2 slots wide, or a GPU that is water cooled which most are even 1 slot.  But I don't think you like or want a water cooled GPU.

George

Proud member of the Old Farts Association

Tom M
Tom M
Joined: 2 Feb 06
Posts: 5,913
Credit: 7,992,285,778
RAC: 6,270,900

https://www.thestreet.com/tec

https://www.thestreet.com/technology-1/two-things-60-minutes-taught-us-about-nvidias-jensen-huang

A Proud member of the O.F.A.  (Old Farts Association).  Be well, do good work, and keep in touch.® (Garrison Keillor)

Tom M
Tom M
Joined: 2 Feb 06
Posts: 5,913
Credit: 7,992,285,778
RAC: 6,270,900

I want to revisit the latest

I want to revisit the latest results of YOUR Over Clocking of your NVIDIA gpus.

I have an EVGA rtx 3080 ti ftw3 ultra I want to push up to the fastest stable processing I can.

Currently about 2/3rds of the e@h brp7/meerKat tasks seem to be running at or below 240s (4 minutes).  This is with no Over Clocking.

A calculation of 4 minutes / task produces a ~1,1M estimate.  If I pick the fastest task times (221s/task), that shifts it up to a ~1.2M.

It seems reasonable that I can speed up this rtx 3080 ti without raising my invalids (much). 

What are your current success stories?

Tom M

A Proud member of the O.F.A.  (Old Farts Association).  Be well, do good work, and keep in touch.® (Garrison Keillor)

Skip Da Shu
Skip Da Shu
Joined: 18 Jan 05
Posts: 145
Credit: 882,959,811
RAC: 2,096,731

Tom,Are setting from a

Tom,

Are setting from a used $420  RTX3080 non-ti I bought last Oct useful to you?

https://einsteinathome.org/host/13168280

# card on top w/ monitor GPU:0(Device0), Gigabyte RTX 3080 10GB

Last look it had 799 valid vs 33 invalid burping meerkats and also runs GPUGRID

Skip

 

PS:  It's sister box with same card -  https://einsteinathome.org/host/13131086

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.