Jump to content

Student building PC for the 1st time for programming

sheenigami

Hi to all! This is my first post, I hope someone of members will help me. I plan to build a deep learning workstation for my school (I am a student) but I have no idea about current hardware and have never build a PC before. What parts should I pick that wouldn't completely destroy my bank. I know for deep learning I require a good GPU, however I do not know any good budget GPU options for deep learning.

 

Link to comment
Share on other sites

Link to post
Share on other sites

Budget and location?

AMD Motherboard Tier List  GPU Cooling Tier List  PSU Tier List  A Dive Into Custom Keyboards & Mechanical Switches

NEWCOMERS Remember to ' Reply ' to comments in order for people to see them, this is done by clicking the arrow icon at the bottom of a comment (Quote).

My Builds:

MEGA Desk Build

 

Blueberry Pi  R9 3950X  Asus X470 ROG Crosshair VII Hero Wi-Fi ATX  Corsair Vengeance RGB Pro 4x16GB 3200MHz CL16  EVGA GTX 1080Ti SC Black Edition 11GB  EVGA P2 850W w/ Blue Sleeved Cables  Cryorig R1 Universal (Blue)  2x Corsair Force MP510 4TB w/ Asus Hyper m.2 V2  Corsair ML Pro Blue LED Fans  Fractal Design Meshify C TG ATX Mid Tower  

Asus ROG SWIFT PG348Q 100Hz IPS G-Sync UW  Dell UltraSharp U3419W 60Hz IPS UW  Custom TOFU96  Corsair Scimitar Pro RGB 

 

Work Rig  ThreadRipper 3970X  Asus Prime TRX40 Pro ATX  G.Skill RipjawsV 8x32GB 3200Mhz CL16  Nvidia Quadro RTX 6000 24GB  Corsair HXi 1000W  EVGA CLC 360 AIO  8x Sabrent Rocket 2TB w/ 2x Asus Hyper m.2 V2   Arctic P12 Fans  Phanteks P400A ATX Mid Tower 

 

Plotting Machine 1  ThreadRipper 2990WX  AsRock X399 Taichi ATX  Kingston HyperX Fury 8x16GB 2666Mhz CL18   Nvidia Quadro K600 1GB  Corsair RMx 850W  EVGA CLC 360 AIO ‖ 2x Sabrent Rocket 2TB & 2x Sabrent Rocket 4TB w/ Asus Hyper m.2 V2  Arctic P12 Fans  Rosewill RSV-L4500 4U  5x Dell DS60 60-bay JBOD w/ 1.2PB of HDD Storage  HP 22U Half Rack

 

Plotting Machine 2  R9 3950X  Asus TUF X570-Plus Wi-Fi ATX  G.Skill RipjawsV 4x16GB 3200Mhz CL16   Nvidia Quadro K600 1GB  Corsair RM 650W  Noctua NH-D15s Chormax ‖ 4x Corsair MP600 Force 1TB W/ Asus Hyper m.2 Gen4  Arctic P14 & P12 Fans  Silverstone FARA R1 ATX Mid Tower

 

J.A.R.V.I.S.  R9 3900XT  Asus B550-I ROG STRIX Wi-Fi ITX  G.Skill Ripjaws V 2x16GB 3600MHz CL16  EVGA GTX 1080 SC2 iCX 8GB  Corsair SF 750W Platinum  Corsair H100i Pro AIO  Noctua NF-A12x15 Chromax Fans  FormD T1 SFF ITX Case  

LG 75UM8070PUA 4K UHD 120Hz IPS HDR TV  Corsair K63 Cherry MX Red Special Edition Wireless  Corsair Ironclaw RGB Wireless

 

Archive Server  R3 2200G  Asus B450-I ROG STRIX Wi-Fi ITX  Corsair Vengeance RGB Pro 2x8GB 3200MHz CL16 White  Vega Integrated Graphics  EVGA P2 750W  Prism Wraith Cooler  ITX Open Bench  2x HP SAS Expander  LSI-SAS9211  220TB of HDD Storage

 

SFF ITX Home PC  i5-7500  MSI B250I Gaming Pro Wi-Fi AC ITX  G.Skill Ripjaws V 2x8GB 2800MHz  Intel Integrated Graphics  Seasonic SSP Flex ATX 300W PSU  Cryorig C7  Velka 3 rev 1.2 SFF ITX Case (Grey)  

 

Laptop  Dell XPS13 2-in-1 7390 (2020)  i7-1065G7  32GB 3733MHz LPDDR4x  Intel Integrated Graphics  1TB NVME M.2  UHD+ (3840 x 2400) InfinityEdge Touch Display 

 

Keyboard Collection  GMMK Full Size  TOFU96 90%  ‖ KBD8x MKII TKL ‖ Drop CTRL High Profile TKL ‖ KBD Bella 75% ‖ GMMK Pro 75% ‖ XD84 Pro 75%  KBD67 V2 MKII 65%  KBD67Lite 65%  TOFU65 65%  KBD Blade 60%  Drop Carina 60%  Southpaw75 60% w/ Left Numpad  OLKB Preonic V3 Ortholinear ‖ CosPad XD24  ‖ KBDPad MKII

 

Key Cap Collection  GMK Boba Fett  GMK Red Samurai  GMK Laser CyberDeck +Novalties  GMK Arch  MaxKey B&W  Drop MT3 Camillo ‖ Matt3o MT3 /dev/tty  KBDfans Biip Torii Ext-2048  ePBT Less But Better +Novalties  EnjoyPBT Dolch  WinMix Mustard  Drop Skylight Horizon & Slate ‖ Glorious Black Aura   

 

Mechanical Key Switch Collection  Zealios V2 65g  Zealios V2 78g ‖ Zilents V2 67g  Tealios V2 67g  C³ Kiwi ‖ C³ Tangerine  Invyr Holy Panda  Durock T1 67g  Kailh Box Thick Jades ‖ Kailh Box Royal ‖ Kailh Box Heavy Dark Yellow ‖ Kailh Box Heavy Burnt Orange  Kailh Box White  Kailh Box Red  Kailh Pro Purple  Kailh Pro Burgandy  Gateron Ink Black V2  Gateron Black  Gateron Yellow  Gateron Brown  Gateron Green  Gateron Blue ‖ Cherry MX Black ‖ Cherry MX Brown ‖ Cherry MX Red

Link to comment
Share on other sites

Link to post
Share on other sites

-= Topic Moved to New Builds and Planning =-

COMMUNITY STANDARDS   |   TECH NEWS POSTING GUIDELINES   |   FORUM STAFF

LTT Folding Users Tips, Tricks and FAQ   |   F@H & BOINC Badge Request   |   F@H Contribution    My Rig   |   Project Steamroller

I am a Moderator, but I am fallible. Discuss or debate with me as you will but please do not argue with me as that will get us nowhere.

 

Spoiler

  

 

Character is like a Tree and Reputation like its Shadow. The Shadow is what we think of it; The Tree is the Real thing.  ~ Abraham Lincoln

Reputation is a Lifetime to create but seconds to destroy.

You have enemies? Good. That means you've stood up for something, sometime in your life.  ~ Winston Churchill

Docendo discimus - "to teach is to learn"

 

 CHRISTIAN MEMBER 

 

 
 
 
 
 
 

 

Link to comment
Share on other sites

Link to post
Share on other sites

4 hours ago, sheenigami said:

Hi to all! This is my first post, I hope someone of members will help me. I plan to build a deep learning workstation for my school (I am a student) but I have no idea about current hardware and have never build a PC before. What parts should I pick that wouldn't completely destroy my bank. I know for deep learning I require a good GPU, however I do not know any good budget GPU options for deep learning.

 

First of all, since you're just starting and you don't seem to have much money to spare, I'd say not to build anything expensive. Since you're still in school, get a cheap laptop and make use of Colab or Kaggle, this way you'll have a powerful GPU for free on the cloud that you're able to use anywhere, this will help you to take your first steps and learn a bit better about ML. When you get acquainted enough to this stuff, you'll probably now what you really need (you don't always need a powerful gpu, sometimes your workloads are CPU-only, or require insane amounts of RAM).

FX6300 @ 4.2GHz | Gigabyte GA-78LMT-USB3 R2 | Hyper 212x | 3x 8GB + 1x 4GB @ 1600MHz | Gigabyte 2060 Super | Corsair CX650M | LG 43UK6520PSA
ASUS X550LN | i5 4210u | 12GB
Lenovo N23 Yoga

Link to comment
Share on other sites

Link to post
Share on other sites

4 hours ago, sheenigami said:

budget GPU

 

4 hours ago, sheenigami said:

deep learning

Pick one.

 

How much is your school willing to spend and what will you be using for deep learning?

Don't ask to ask, just ask... please 🤨

sudo chmod -R 000 /*

Link to comment
Share on other sites

Link to post
Share on other sites

If you're on a Budget, there is no way you can get started with deep learning on your local GPU. The majority of DeepLearning frameworks, especially the more popular ones, only support CUDA for GPU acceleration, which means you'd have to get a NVIDIA card. NVIDIA cards tend to be more expensive than AMD cards.

 

I'd recommend buying a notebook or a budget friendly PC without paying too much attention to the GPU, since budget friendly GPUs won't help you a lot with deep learning. Small deep learning models will train just fine if you get a decent CPU, for larger models you can use Cloud offerings like Google's Colab. If you make progress and have more money to spend, you can still opt for higher end NVIDIA Cards or Cloud instances with GPUs for trainings.

 

Link to comment
Share on other sites

Link to post
Share on other sites

Sorry for the late reply, I am new this form and thought replies would get sent to my email.

On 2/11/2020 at 2:07 PM, JM21 said:

Budget and location?

Anyways, I live in Singapore and for budget I am looking at around $2k SGD which translates to about 1,442.57
USD and 1,914.92 CAD.

17 hours ago, MoVo said:

Small deep learning models will train just fine if you get a decent CPU, for larger models you can use Cloud offerings like Google's Colab. If you make progress and have more money to spend, you can still opt for higher end NVIDIA Cards or Cloud instances with GPUs for trainings.

 

 I have used google collab in the past before, but i did not know it could be used for deep learning, maybe i should just stick with google collab, depending on what kind of computer i am able to build with my budget.

22 hours ago, igormp said:

When you get acquainted enough to this stuff, you'll probably now what you really need (you don't always need a powerful gpu, sometimes your workloads are CPU-only, or require insane amounts of RAM).

Currently I already own a laptop for school (Asus Ux331UN Ram: 8GB, CPU: i7-8550U, GPU: MX150, SSD: 256 GB), however, my lecturer in school told me that I would need a decent computer for deep learning if not I would not be able to do my assignment or it will take my laptop very long in to run.

 

Sorry again for the late reply,

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, sheenigami said:

Sorry for the late reply, I am new this form and thought replies would get sent to my email.

Anyways, I live in Singapore and for budget I am looking at around $2k SGD which translates to about 1,442.57
USD and 1,914.92 CAD.

 I have used google collab in the past before, but i did not know it could be used for deep learning, maybe i should just stick with google collab, depending on what kind of computer i am able to build with my budget.

Currently I already own a laptop for school (Asus Ux331UN Ram: 8GB, CPU: i7-8550U, GPU: MX150, SSD: 256 GB), however, my lecturer in school told me that I would need a decent computer for deep learning if not I would not be able to do my assignment or it will take my laptop very long in to run.

 

Sorry again for the late reply,

Your lecturer is an asshole tbh. Colab sure is more than enough for most starter tasks in deep learning (heck, it has tons of vram), and even your laptop would be good enough to do so. I've trained some models using my laptop's 840m without much trouble (took around 4 hours to train stuff), and it already was many times faster than doing so with a CPU. 

 

Again, I would only recommend you to buy new stuff when you actually feel hindered by your current setup, and by then you'd already know what you specifically need. Also, by then new GPUs would be released with way better ML performance, so it's a win win anyway.

FX6300 @ 4.2GHz | Gigabyte GA-78LMT-USB3 R2 | Hyper 212x | 3x 8GB + 1x 4GB @ 1600MHz | Gigabyte 2060 Super | Corsair CX650M | LG 43UK6520PSA
ASUS X550LN | i5 4210u | 12GB
Lenovo N23 Yoga

Link to comment
Share on other sites

Link to post
Share on other sites

22 hours ago, igormp said:

Your lecturer is an asshole tbh. Colab sure is more than enough for most starter tasks in deep learning (heck, it has tons of vram), and even your laptop would be good enough to do so. I've trained some models using my laptop's 840m without much trouble (took around 4 hours to train stuff), and it already was many times faster than doing so with a CPU. 

 

Again, I would only recommend you to buy new stuff when you actually feel hindered by your current setup, and by then you'd already know what you specifically need. Also, by then new GPUs would be released with way better ML performance, so it's a win win anyway.

Ahhhhhhhhhhh ok. Thanks for your help, i think i would put off building a pc right now

Link to comment
Share on other sites

Link to post
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

×