Jump to content

Welcome to ExtremeHW

Welcome to ExtremeHW, register to take part in our community, don't worry this is a simple FREE process that requires minimal information for you to signup.

 

Registered users can: 

  • Start new topics and reply to others.
  • Show off your PC using our Rig Creator feature.
  • Subscribe to topics and forums to get updates.
  • Get your own profile page to customize.
  • Send personal messages to other members.
  • Take advantage of site exclusive features.
  • Upgrade to Premium to unlock additional sites features.
IGNORED

Modder creates method to run AI on cheap AMD APUs


UltraMega

Recommended Posts

Quote

The 4600G is currently selling at price of $95. It includes a 6-core CPU and 7-core GPU. 5600G is also inexpensive — around $130 with better CPU but the same GPU as 4600G.
 

It can be turned into a 16GB VRAM GPU under Linux and works similar to AMD discrete GPU such as 5700XT, 6700XT, …. It thus supports AMD software stack: ROCm. Thus it supports Pytorch, Tensorflow. You can run most of the AI applications.

16GB VRAM is also a big deal, as it beats most of discrete GPU. Even those GPU has better computing power, they will get out of memory errors if application requires 12 or more GB of VRAM. Although the speed is an issue, it’s better than out of memory errors.

 

Democratize AI: turn a $95 chip into a 16GB VRAM GPU! Beats most of the discrete GPUs! | by Ttiotech | Aug, 2023 | Medium

 

TLDR: AI needs a decent amount of Vram to work. GPUs with enough Vram are expensive. This is a way around that problem. 

 

  • Respect 1

null

Owned

 Share

CPU: 5800x
MOTHERBOARD: ASUS TUF Gaming B550-Plus
RAM: 32GB 3600mhz CL16
GPU: 7900XT
SOUNDCARD: Sound Blaster Z 5.1 home theater
MONITOR: 4K 65 inch TV
Full Rig Info
Link to comment
Share on other sites

Wow, talk about a budget workaround!

£3000

Owned

 Share

CPU: AMD Ryzen 9 7950X3D
MOTHERBOARD: MSI Meg Ace X670E
RAM: Corsair Dominator Titanium 64GB (6000MT/s)
GPU: EVGA 3090 FTW Ultra Gaming
SSD/NVME: Corsair MP700 Pro SE Gen 5 4TB
PSU: EVGA Supernova T2 1600Watt
CASE: be quiet Dark Base Pro 900 Rev 2
FANS: Noctua NF-A14 industrialPPC x 6
Full Rig Info

Owned

 Share

CPU: Intel Core i5 8500
RAM: 16GB (2x8GB) Kingston 2666Mhz
SSD/NVME: 256GB Samsung NVMe
NETWORK: HP 561T 10Gbe (Intel X540 T2)
MOTHERBOARD: Proprietry
GPU: Intel UHD Graphics 630
PSU: 90Watt
CASE: HP EliteDesk 800 G4 SFF
Full Rig Info

£3000

Owned

 Share

CPU: 2 x Xeon|E5-2696-V4 (44C/88T)
RAM: 128GB|16 x 8GB - DDR4 2400MHz (2Rx8)
MOTHERBOARD: HP Z840|Intel C612 Chipset
GPU: Nvidia Quadro P2200
HDD: 4x 16TB Toshiba MG08ACA16TE Enterprise
SSD/NVME: Intel 512GB 670p NVMe (Main OS)
SSD/NVME 2: 2x WD RED 1TB NVMe (VM's)
SSD/NVME 3: 2x Seagate FireCuda 1TB SSD's (Apps)
Full Rig Info
Link to comment
Share on other sites

In other news: 

WWW.TECHSPOT.COM

For every H100 GPU accelerator sold, Nvidia appears to be making a remarkable profit, with reported margins reaching 1,000 percent of production costs. Tae Kim, a senior...

 

null

Owned

 Share

CPU: 5800x
MOTHERBOARD: ASUS TUF Gaming B550-Plus
RAM: 32GB 3600mhz CL16
GPU: 7900XT
SOUNDCARD: Sound Blaster Z 5.1 home theater
MONITOR: 4K 65 inch TV
Full Rig Info
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×
×
  • Create New...

Important Information

This Website may place and access certain Cookies on your computer. ExtremeHW uses Cookies to improve your experience of using the Website and to improve our range of products and services. ExtremeHW has carefully chosen these Cookies and has taken steps to ensure that your privacy is protected and respected at all times. All Cookies used by this Website are used in accordance with current UK and EU Cookie Law. For more information please see our Privacy Policy