Jump to content

Budget (including currency): 91000 INR ($1096 Approx)

Country: India 

Games, programs or workloads that it will be used for: For Programming including AI and Machine Learning (beginner level), Valorant, Forza Horizon, Blender, and COD Warzone. (All at 1080P)

Other details (existing parts lists, whether any peripherals are needed, what you're upgrading from, when you're going to buy, what resolution and refresh rate you want to play at, etc): Gonna Buy a PC next month with the specifications that is attached, budget kinda tight but do need the i5-12600K, and the DDR5 Ram. Have all the Peripherals from my current setup which is a 2015 PC with an i3 3rd Gen and GT610 GPU.

There are TWO similar Configurations that I have made, one is a little cost cut and other is not (according to me)

 

 

i5-12600k RTX 3050 83.5k.pdf i5-12600K RTX3050 MSI.pdf

Link to comment
https://linustechtips.com/topic/1537407-recommendations-needed-for-pc-build/
Share on other sites

Link to post
Share on other sites

With all the stuff you're going to do, I rlly recommend at least 32GB of ram, and a better cpu/gpu combo, I recommend saving for a 3060, it has 12GB of vram instead of 8 like 3050, so it will do way better in AI, either get a cheaper cpu or save up a bit more, that's just way better value

Link to post
Share on other sites

I don’t know the used/new market in India much but maybe swapping out the 3050 for an RX 6600 as that will perform closer and in some cases better than a 3060 for the same price as the 3050. (At least in the U.S)

Link to post
Share on other sites

1 hour ago, MiszS said:

With all the stuff you're going to do, I rlly recommend at least 32GB of ram, and a better cpu/gpu combo, I recommend saving for a 3060, it has 12GB of vram instead of 8 like 3050, so it will do way better in AI, either get a cheaper cpu or save up a bit more, that's just way better value

You mean this could work? 

i5-13400F RTX 3060 12GB All MSI.pdf

Link to post
Share on other sites

4 hours ago, Babylion122 said:

I don’t know the used/new market in India much but maybe swapping out the 3050 for an RX 6600 as that will perform closer and in some cases better than a 3060 for the same price as the 3050. (At least in the U.S)

AMD GPUs are a bad pick for machine learning.

 

+1 on the 3060 idea.

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 post
Share on other sites

4 hours ago, Danish Jamdar said:

yeah it seems like a good pc, but make sure to upgrade to 32GB of ram later, i don't know the exact prices for you, but make sure to get the cheapest 3060, because the different brands rlly don't matter for the perfomance. Also make sure to get the thermalright cpu contact frame, this way the cpu won't bend, and you will have better temps

Link to post
Share on other sites

10 hours ago, MiszS said:

yeah it seems like a good pc, but make sure to upgrade to 32GB of ram later, i don't know the exact prices for you, but make sure to get the cheapest 3060, because the different brands rlly don't matter for the perfomance. Also make sure to get the thermalright cpu contact frame, this way the cpu won't bend, and you will have better temps

Yeah the 3060's here are only a tad bit costly ($48) so can do and the ram I'm probably gonna upgrade in less than a year, so that too. Will have to see the contact frame with the shop I'm going to build with, so thanks, I have actually heard about that earlier this week. But is Deepcool AG620 good?

Link to post
Share on other sites

For machine learning we'd need to know basically everything. So resource intensive the slightest bit off could slow down your projects.

 

I'd strongly recommend looking at the used market cause $1000 is low already for light stuff and to dive into the deep end demanding stuff will be a fight.

In my mind AI stuff is heavy on core count, try and get something like a used Intel i7-11700F or i9-10850K for them cores at a low price. From the AI stuff I've tried they're very storage and GPU dependent. 3060 12GB is the only relevant GPU here that shouldn't cost and arm and a leg. Lots of RAM, 32GB minimum and get an SSD with tons of storage.

 

I would strongly recommend an external UPS. Losing the last 30 minutes of a word document can be annoying because of a power outage. Loosing the last 12 hours of computation because of a power outage would be something I'd find hard to recover from.

MAIN: Ryzen 7 5800X3D - Kraken X62 Rev 2 - STRIX X470-I - 3600MHz 32GB Kingston Fury - 250GB 970 Evo boot - 2x 500GB 860 Evo - 1TB P3 - 4TB HDD - RX6800 - Antec HCG Platinum - Manta - Silent Wings Pro 4's enjoyer

SetupZowie XL2740 27.0" 240hz - Roccat Burt Pro OG Corsair K70 browns - PC38X - Mackie CR5X's Mackie CR8S-XBT

Current build on PCPartPicker

 

 

HTPC: Ryzen 7 2700X - BeQuiet! Shadow Rock 3 - STRIX X570-F - 3200MHz 32GB Corsair Dominator - 250GB Exceria boot - 500GB SN730 - 1TB Sandisk 3D - 4TB HDD - Limited Edition Vega 64 - Corsair RM750x 80+ Gold - North - Alphacool Apex Stealth Metal - BeQuiet! Light Wings

SetupHisense 55E7NQ - Hisense HS205G

HTPC on PCPartPicker

Link to post
Share on other sites

4 hours ago, venomtail said:

For machine learning we'd need to know basically everything. So resource intensive the slightest bit off could slow down your projects.

 

I'd strongly recommend looking at the used market cause $1000 is low already for light stuff and to dive into the deep end demanding stuff will be a fight.

In my mind AI stuff is heavy on core count, try and get something like a used Intel i7-11700F or i9-10850K for them cores at a low price. From the AI stuff I've tried they're very storage and GPU dependent. 3060 12GB is the only relevant GPU here that shouldn't cost and arm and a leg. Lots of RAM, 32GB minimum and get an SSD with tons of storage.

 

I would strongly recommend an external UPS. Losing the last 30 minutes of a word document can be annoying because of a power outage. Loosing the last 12 hours of computation because of a power outage would be something I'd find hard to recover from.

Actually I will be just starting the programming classes 1st half of the next year so I have plenty of time to upgrade it. But the i5-12600K is better than i7-11700K, so I think I can stick with it, and try to adjust a RTX 3060 in the build. As for the Ram its quite easily upgradable so i can do that after a few months of the Build. I already have 2 TB HDD and 320GB HDD and a 512GB Sata SSD, so storage is no issue and already have like 3 UPS lying around. So thanks for the recommendations. I'll just see to get any and every discount possible to get the RTX 3060 and another M.2.

Link to post
Share on other sites

17 hours ago, Danish Jamdar said:

Yeah the 3060's here are only a tad bit costly ($48) so can do and the ram I'm probably gonna upgrade in less than a year, so that too. Will have to see the contact frame with the shop I'm going to build with, so thanks, I have actually heard about that earlier this week. But is Deepcool AG620 good?

Yeah it's a pretty good cooler, the shop most likely won't have the contact frame, so you should order it online

Link to post
Share on other sites

On 10/25/2023 at 5:17 AM, venomtail said:

In my mind AI stuff is heavy on core count, try and get something like a used Intel i7-11700F or i9-10850K for them cores at a low price.

Not really, having a beefier GPU is always better. If the build is solely made for ML and you want to minmax it, an i3 with a 4090 would be preferable.

On 10/25/2023 at 5:17 AM, venomtail said:

From the AI stuff I've tried they're very storage and GPU dependent.

GPU and RAM, actually. Storage is mostly for when you don't have enough memory and want to do some off-core processing, which won't be the case for someone who's getting started.

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 post
Share on other sites

12 hours ago, igormp said:

Storage is mostly for when you don't have enough memory and want to do some off-core processing

Correct, that's the thing. At the time I only had 16GB and I spilled like 100GB onto my HDD (god it was so slow) and that was before I realised what pagefiles were and that they were on my HDD, so I came to the conclusion GPU and Storage. Now it's clicked cause I've realised what I didn't realise then.

MAIN: Ryzen 7 5800X3D - Kraken X62 Rev 2 - STRIX X470-I - 3600MHz 32GB Kingston Fury - 250GB 970 Evo boot - 2x 500GB 860 Evo - 1TB P3 - 4TB HDD - RX6800 - Antec HCG Platinum - Manta - Silent Wings Pro 4's enjoyer

SetupZowie XL2740 27.0" 240hz - Roccat Burt Pro OG Corsair K70 browns - PC38X - Mackie CR5X's Mackie CR8S-XBT

Current build on PCPartPicker

 

 

HTPC: Ryzen 7 2700X - BeQuiet! Shadow Rock 3 - STRIX X570-F - 3200MHz 32GB Corsair Dominator - 250GB Exceria boot - 500GB SN730 - 1TB Sandisk 3D - 4TB HDD - Limited Edition Vega 64 - Corsair RM750x 80+ Gold - North - Alphacool Apex Stealth Metal - BeQuiet! Light Wings

SetupHisense 55E7NQ - Hisense HS205G

HTPC on PCPartPicker

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

×