Jump to content
To encourage social distancing, you must leave two blank lines at the start and end of every post, and before and after every quote. Failure to comply may result in non-essential parts of the forum closing. Click for more details. ×
Search In
  • More options...
Find results that contain...
Find results in...
ProgrammingDude

Building a workstation for programming and work

Recommended Posts

Posted · Original PosterOP

I am looking for a desktop pc that is primarily used for using Excel, 1-2 JetBrains IDE (includes running local workloads from time to time for testing), Chrome and other minor programs. It also runs (i.e. connects to) two virtual desktops.

I am not doing any Gaming or Video rendering. But I do have 4 monitors at the moment, 2 of them are 4K already (LG 32UD99-W), but I plan to replace the other two with similar monitors as well (suggestions, please, but don't count towards the budget). It needs to be an improvement over my current HP EliteBook 850 G5 (15.60", UHD, Intel Core i7-8550U, 16GB, 1000GB, SSD, 4G) which I find to be too slow.

 

PC should be able to run 1 or 2 8K monitors in the future, ideally.

I also have a mouse and a keyboard already. It should be as white as possible. The CPU cooler is not listed in white on the PCPartPicker website, but is available.

 

Here is my suggestion:

PCPartPicker Part List

CPU: AMD Ryzen 7 3700X 3.6 GHz 8-Core Processor  (€336.89 @ Mindfactory) 
CPU Cooler: Alpenföhn Brocken 3 60.62 CFM CPU Cooler  (€51.98 @ Alternate) 
Motherboard: Asus PRIME X570-P ATX AM4 Motherboard  (€182.90 @ Alternate) 
Memory: Crucial Ballistix Sport LT 32 GB (2 x 16 GB) DDR4-3200 Memory  (€161.99 @ Amazon Deutschland) 
Memory: Crucial Ballistix Sport LT 32 GB (2 x 16 GB) DDR4-3200 Memory  (€161.99 @ Amazon Deutschland) 
Storage: Western Digital SN750 1 TB M.2-2280 NVME Solid State Drive  (€181.29 @ Amazon Deutschland) 
Video Card: Gigabyte GeForce RTX 2070 SUPER 8 GB Gaming OC 3X Video Card  (€553.79 @ Mindfactory) 
Case: Fractal Design Meshify C ATX Mid Tower Case  (€96.89 @ Alternate) 
Power Supply: be quiet! Straight Power 11 650 W 80+ Gold Certified Fully Modular ATX Power Supply  (€106.99 @ Amazon Deutschland) 
Case Fan: ARCTIC P12 PWM PST 56.3 CFM 120 mm Fan  (€6.37 @ Amazon Deutschland) 
Case Fan: ARCTIC P12 PWM PST 56.3 CFM 120 mm Fan  (€6.37 @ Amazon Deutschland) 
Case Fan: ARCTIC P12 PWM PST 56.3 CFM 120 mm Fan  (€6.37 @ Amazon Deutschland) 
Total: €1853.82
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2020-01-19 17:59 CET+0100

 

Is this a reasonable build? I want something that will last me ideally five years or potentially more.

 

Thank you very much!

Link to post
Share on other sites

Overkill, possibly.  Programming needs nothing. People program on Pis.  An Intel iGPU will drive 4 4K monitors @60hz.  The question is the other programs.  I’ve heard of really complicated excel stuff liking a lot of threads, and I know not a thing about jetbrains.  If none of it uses CUDA cores though the big GPU is kinda wasted.  It doesn’t have any more video memory than a 570 8gig.


Life is like a bowl of chocolates: there are all these little crinkly paper cups everywhere.

Link to post
Share on other sites
Posted · Original PosterOP

Jetbrains IDEs are a memory eater, they require a boatload. On my current setup, they are always lagging.

Excel is the same thing for large data work. I use Excel often to modify data, transform it and what not and running larger of those sheets takes a lot of time with my current computer. And it triggers me when entering something into a field has a 0.7sec lag, it just isn't smooth.

 

The GPU is potentially an overkill, but I might want to do some video editing in the future or local ML. If this is the only overkill, I'd be happy to live with it.

Link to post
Share on other sites
34 minutes ago, ProgrammingDude said:

Jetbrains IDEs are a memory eater, they require a boatload. On my current setup, they are always lagging.

Excel is the same thing for large data work. I use Excel often to modify data, transform it and what not and running larger of those sheets takes a lot of time with my current computer. And it triggers me when entering something into a field has a 0.7sec lag, it just isn't smooth.

 

The GPU is potentially an overkill, but I might want to do some video editing in the future or local ML. If this is the only overkill, I'd be happy to live with it.

Whelp if that’s the case you could up the ram to 64gb, add 4 more threads with a 3900x, do a Radeon 570 8gb and do it while save money.  Or you could enjoy your video games ;) 


Life is like a bowl of chocolates: there are all these little crinkly paper cups everywhere.

Link to post
Share on other sites

I have a very similar build and its great I would suggest added an Nvme SSD for a boot drive instead of another 32Gb of ram. I have 2x for windows and linux. 

47 minutes ago, Bombastinator said:

Jetbrains IDEs are a memory eater

Yes, this is a problem but I have have never hit by limit with both Pycharm, Matlab and Chrome open. usually hover around 16Gb off usage.

Link to post
Share on other sites
1 hour ago, NaGoldfarb said:

I have a very similar build and its great I would suggest added an Nvme SSD for a boot drive instead of another 32Gb of ram. I have 2x for windows and linux. 

Yes, this is a problem but I have have never hit by limit with both Pycharm, Matlab and Chrome open. usually hover around 16Gb off usage.

I did it say this.  The OP did.  I did quote his statement.  I don’t know how it got attributed to me.  I know nothing about jetbrains.  


Life is like a bowl of chocolates: there are all these little crinkly paper cups everywhere.

Link to post
Share on other sites
3 hours ago, ProgrammingDude said:

Jetbrains IDEs are a memory eater, they require a boatload. On my current setup, they are always lagging.

Excel is the same thing for large data work. I use Excel often to modify data, transform it and what not and running larger of those sheets takes a lot of time with my current computer. And it triggers me when entering something into a field has a 0.7sec lag, it just isn't smooth.

 

The GPU is potentially an overkill, but I might want to do some video editing in the future or local ML. If this is the only overkill, I'd be happy to live with it.

The GPU will also help with if you tensor flow or simulations.

Link to post
Share on other sites
6 minutes ago, NaGoldfarb said:

The GPU will also help with if you tensor flow or simulations.

Point.  If you can actually make use of CUDA an Nvidia GPU will help a lot.  The thing is basically a collection of specialized floating point math coprocessors.  They can be pretty specialized though and things have to be written to use them.


Life is like a bowl of chocolates: there are all these little crinkly paper cups everywhere.

Link to post
Share on other sites
Posted · Original PosterOP
3 hours ago, Bombastinator said:

Whelp if that’s the case you could up the ram to 64gb, add 4 more threads with a 3900x, do a Radeon 570 8gb and do it while save money.  Or you could enjoy your video games ;) 

I do have 64GB already, do you reckon the Radeon + 3900X would be better for me? What about something like a 1660 Super?

3 hours ago, NaGoldfarb said:

I have a very similar build and its great I would suggest added an Nvme SSD for a boot drive instead of another 32Gb of ram. I have 2x for windows and linux. 

Yes, this is a problem but I have have never hit by limit with both Pycharm, Matlab and Chrome open. usually hover around 16Gb off usage.

I ran out of memory yesterday again. I have 16GB at the moment.

Do you have a recommendation for the NVME SSD boot drive?

40 minutes ago, NaGoldfarb said:

The GPU will also help with if you tensor flow or simulations.

Simulations are another thing I've missed. I run a bunch of them as it is today. Usually go out of the room while they are running.

31 minutes ago, Bombastinator said:

Point.  If you can actually make use of CUDA an Nvidia GPU will help a lot.  The thing is basically a collection of specialized floating point math coprocessors.  They can be pretty specialized though and things have to be written to use them.

I think you have a point with your remarks, I am just not sure if I won't use the GPU once I have it. But see my question above.

Link to post
Share on other sites
17 hours ago, ProgrammingDude said:

do you reckon the Radeon

I would not recommend a AMD graphics if you want to do ML. They do not have CUDA, which most ML libraries can use.

Link to post
Share on other sites

So it sounds like the graphics card may help a lot after all.  The question then becomes which one.  A 2070s is a 2070 with gddr6 instead of gddr5.  If CUDA core count is what will make a difference might a 1080 actually be faster for this use case? It lacks rtx stuff and ddr6, but it’s got more CUDA cores I think.


Life is like a bowl of chocolates: there are all these little crinkly paper cups everywhere.

Link to post
Share on other sites
23 hours ago, Bombastinator said:

Programming needs nothing.

If one earns a living writing software this approach is sure to reduce income.

 

@ProgrammingDude,

 

If you want 64GB of memory consider using a 2 x  32GB kit. This would leave two free slots for an upgrade if needed. 

 

If you go with 16GB memory modules get a 4x16GB kit. XMP profiles are based on the modules in the kit. Getting two kits might require some manual adjustments to achieve advertised speeds.

 

There should be no need for the add-on fans.

 

Consider a Corsair RMX or Seasonic Prime psu.


80+ ratings certify electrical efficiency. Not quality.

 

Link to post
Share on other sites
Just now, brob said:

If one earns a living writing software this approach is sure to reduce income.

 

@ProgrammingDude,

 

If you want 64GB of memory consider using a 2 x  32GB kit. This would leave two free slots for an upgrade if needed. 

 

If you go with 16GB memory modules get a 4x16GB kit. XMP profiles are based on the modules in the kit. Getting two kits might require some manual adjustments to achieve advertised speeds.

 

There should be no need for the add-on fans.

 

Consider a Corsair RMX or Seasonic Prime psu.

Needs and wants are a bit different.  Coding and compiling are different things.  For coding the slowest dual core around is overkill.  It’s just a text editor.  Not even a word processor.  That ship has sailed though because apparently CUDA cores are on the menu.


Life is like a bowl of chocolates: there are all these little crinkly paper cups everywhere.

Link to post
Share on other sites
Posted · Original PosterOP
1 hour ago, brob said:

If one earns a living writing software this approach is sure to reduce income.

 

@ProgrammingDude,

 

If you want 64GB of memory consider using a 2 x  32GB kit. This would leave two free slots for an upgrade if needed. 

 

If you go with 16GB memory modules get a 4x16GB kit. XMP profiles are based on the modules in the kit. Getting two kits might require some manual adjustments to achieve advertised speeds.

 

There should be no need for the add-on fans.

 

Consider a Corsair RMX or Seasonic Prime psu.

I was a bit unsure about the add-on fans, so I will potentially leave them out.

I had an RMX in there then someone recommended the be quiet!

What makes you choose RMX over be quiet?

 

I think I'll stick to the 2×2×16GB because the 1×4×16GB cost 25% more for whatever reason and 32GB modules are not available.

If I require an update at a later stage, I can probably just replace 2×16GB and be done with it, right?

Link to post
Share on other sites
24 minutes ago, ProgrammingDude said:

I was a bit unsure about the add-on fans, so I will potentially leave them out.

I had an RMX in there then someone recommended the be quiet!

What makes you choose RMX over be quiet?

 

I think I'll stick to the 2×2×16GB because the 1×4×16GB cost 25% more for whatever reason and 32GB modules are not available.

If I require an update at a later stage, I can probably just replace 2×16GB and be done with it, right?

The PSU tierlist.  PSU quality is mostly about service life, resistance to uneven wall voltage, and the ability to the PSU to not pass on system destroying voltage levels if(when) it fails.
 

 


Life is like a bowl of chocolates: there are all these little crinkly paper cups everywhere.

Link to post
Share on other sites
4 hours ago, ProgrammingDude said:

What makes you choose RMX over be quiet?

i believe that RMX are better units than Straight Power 11. RMX are quieter and have much better warranties. While I don't generally factor warranty in a psu decision, when one manufacturer offers twice the warranty, (10 years vs 5), it reinforces my opinion of component quality. Also, I'm not a fan of multi rail psu.


80+ ratings certify electrical efficiency. Not quality.

 

Link to post
Share on other sites
Posted · Original PosterOP

I have one more question on top of the question about the RAM earlier:

11 hours ago, ProgrammingDude said:

Further question: is a 10% more expensive G.Skill Trident Z Neo 32 GB (2 x 16 GB) DDR4-3600 Memory worth the upgrade to 3600?

 

I could get the Gigabyte GeForce RTX 2070 SUPER 8 GB Gaming OC 3X for 535€ or the Gigabyte GeForce Rtx 2060 Super Gaming OC white 8GD for 419€, i.e. 21.6% cheaper (116€ in absolute terms). Could anyone give me a recommendation one way or the other?

Link to post
Share on other sites
2 hours ago, ProgrammingDude said:

I have one more question on top of the question about the RAM earlier:

 

I could get the Gigabyte GeForce RTX 2070 SUPER 8 GB Gaming OC 3X for 535€ or the Gigabyte GeForce Rtx 2060 Super Gaming OC white 8GD for 419€, i.e. 21.6% cheaper (116€ in absolute terms). Could anyone give me a recommendation one way or the other?

i'd get the faster ram. if you don't game and you don't think you'll have huge 3d projects or huge video editing projects, go for the 2060 super.

the gpu is the easiest component to upgade, if necessary.

Link to post
Share on other sites

Faster RAM will help a lot with a Ryzen CPU. The GPU is defiantly the easiest to upgrade. The 2060 super is a great card, the 2070 super is probably over kill. 

Link to post
Share on other sites

I can power a screen shot later with of my CPU/RAM usage under a work load you might have, an instance or two of pycharm, a browser with a dozen or two tabs open. it will be in linux but should still give you an idea of what you can expect.

 

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


×