lyinch

That's a lot of servers... Looking trough the endless web I found some requirements for the different game servers Rust: min 4GB RAM, 8GB+ recommended Arma: 2GB RAM, 4GB+ recommended Ark: 8GB RAM, 16GB+ recommended (crazy) CS Go: 2GB RAM, 6GB+ recommended Since you mentioned many users, you can scale up those numbers. If you want to run all those servers concurrently, I would recommend 16GB-32GB of RAM and at least one dedicated core per instance. I don't know if the servers parallelise well. It might therefore make more sense to have better per-core performance rather than many cores. Maybe you have some experience there if you're already running some servers. Don't forget about a good networking card and SSDs. Besides just the HW requirements, I advise you to look into virtualisation to limit the resources per process such that a roque server doesn't impact the others. If you don't have any experience with server hosting, rent a cheap VPS then host each game server sequentially for some time and look at their requirements.

That's more than enough. I did an internship as a game developer (using Unity) and worked with a 2015 Macbook Pro 16GB RAM 1867 MHzm 3.1GHz i7, and the build-in GPU. If you have no experience at all then you're fighting the books more than your machine. By the time you reach the limits of your gaming rig you know what your specific needs are. Who knows, maybe this is the start of your programming career? Or the beginnings of your 3D modelling career? They require fundamental different setups and until you've done a few small projects and your plan gets more concrete there is no need to upgrade. You don't need a powerful machine to learn programming. Besides the PC build, I recommend that you start with 2D games before you dive in the world of quaternions.

That is very interesting to hear, thank you for the two links. I really like the reports of pugetsystems, that's also were I got the confirmation of thermal throttling for multi-GPU setups. To wrap this up, I decided to go with the build. As advised by @Juular I chose a different case, the Phanteks Enthoo Pro. The other changes were the CPU aio cooler as it was not available in Switzerland, the RAM which wasn't available either, and the HDD where I went from the 4TB model to the 3TB WD Red HDD because they are surprisingly expensive here. The major components stayed the same, I chose the cheapest blower GPU that I could find, which was in my case the Zotac RTX 2080Ti for 1098CHF or $1123 and sticked with the motherboard, CPU, PSU and NVME. For anyone interested, if you take the lowest functional subset of components, namely no HDDs and "only" one GPU, the build costs (+shipping etc.) 3074CHF or 2872€ or $3144. Adding the GPU and the hard disks raises the price by 1304CHF to an insane 4.3kCHF. Let's not think about the two free GPU slots... I'm excited to build this and hope that everything not only fits together but also works well together. Thank you all for your great support. The next post here will probably be a request for a custom water loop!

Correct, looking at the docs for e.g. tensorflow it is almost trivial to add more GPUs. However, it's not always so easy use the full power of both GPUs (might not get a 2x speedup, but rather 1.7x or something) because some tasks have a heavy synchronization overhead. Adding an NVLink bridge might help but I haven't looked into this, yet. I assume it's easy to add once I have the GPUs, or is this something that I need to plan in advance? In general, the 2080Ti and Titan RTX are both amazing cards for this task and I think that I won't suffer by choosing the slightly cheaper setup. For those interested, the two GPUs don't need to be used in parallel for the exact same model. It is possible to train the model with different hyperparameters on each to "really get the full 2x speedup".

I can't comment on the specific models of each component but the overall build looks good. 32GB of RAM is not exaggerated here as IDEs use quite a lot of memory for full code coverage, and running a bunch of docker containers eats it up faster than you can free it. We're not even speaking about the consumption of electron apps or just the gazillion tabs that you have open. The multicore CPU is a very good choice if you have long compile times such as Android development, large scala or c++ code bases or heavily use virtualisation, otherwise 8 cores aren't really needed. I would definitely go for a 500GB SSD and put all your code on this as it speeds up file indexing and tracking tremendously. Remember also that your docker images cache can take up a significant amount of space. If you have good file management then you might make it work with 240GB and offload unused files to a cold storage such as a standard harddisk, but this is more cumbersome. You can cheap out on the GPU as a normal developer if you don't touch it explicitly. Remember that webgl also benefits from a good GPU if you're roaming around in the javascript world which is common for a SAAS developer.

I agree with the case. I didn't put much thought into it and it's the only part that I chose because it looks good Thank you for the extensive list of cases, I'll take my time tomorrow and go through them. You hit all the pain points, the CPU cooler is the other part that I had on the list and never looked at again. I'll follow your guidance here and chose the EVGA CLC 360. Please ignore the prices of the components on my list, I have to buy them in Switzerland and pcpp doesn't support it. I looked some parts up at local retailers and they differ by a lot (in both directions). I can get a PNY 2080 Ti for 1093CHF and the ZOTAC 2080 Ti (which is in my build) for 1069CHF. That's why I added this specific card to the build. As for the SSD, I don't have any fixed opinion either. If I can save some money there, then I'll probably use the slightly cheaper one. (edit:) I'll be working in this field but probably won't make money with it as a side gig. So the machine is strictly personal and not professional. Although it helps me for my studies and job, it's not required for my main income. Thank you all for taking the time to give me feedback!

Pricewise the 7920x is twice that of the Threadripper. I could get one for 1.1k but the above mentioned Threadripper for 500. The Intel CPU has "only" 44PCIe lanes. Am I too focused on the 3 x16 + 1 x8 lanes here? I haven't considered the Titan V, that will not fit my budget. I'm already stressing it enough with this build. I want a (very) powerful machine at home, and anything that needs more computational power I'll offload it to the cloud. I did consider the Titan RTX, but I can get two 2080Ti for 2.2k which are blower style, or 2.8k with watercooled aio whereas the Titan costs 3k here and has dual fans which won't cooperate with any future additions.

In this case it would be an nVidia/AMD build. What's the equivalent Intel CPU? Do I need to look into Xeon CPUs for something that supports the equivalent PCIe lanes than the 2nd gen Threadripper? I'm at a loss when looking into the convoluted world of Intel processors.

Hey, I plan to build a (multi-GPU) deep learning workstation. It should be upgradable (replace individual components) and extendable (add components). I have no idea about current hardware trends so feel free to take it all apart. I need prices for Switzerland. PCPartPicker Part List CPU: AMD Threadripper 2920X 3.5 GHz 12-Core Processor ($449.98 @ Amazon) CPU Cooler: Deepcool CASTLE 240EX 64.4 CFM Liquid CPU Cooler ($120.99 @ Amazon) Motherboard: Asus ROG ZENITH EXTREME EATX sTR4 Motherboard Memory: G.Skill Ripjaws V Series 64 GB (4 x 16 GB) DDR4-3200 Memory ($266.99 @ Newegg) Storage: ADATA XPG SX8200 Pro 1 TB M.2-2280 NVME Solid State Drive ($149.99 @ Amazon) Storage: Western Digital Red 4 TB 3.5" 5400RPM Internal Hard Drive ($116.99 @ Adorama) Storage: Western Digital Red 4 TB 3.5" 5400RPM Internal Hard Drive ($116.99 @ Adorama) Video Card: Zotac GeForce RTX 2080 Ti 11 GB Blower Video Card (2-Way SLI) ($1612.98 @ Amazon) Video Card: Zotac GeForce RTX 2080 Ti 11 GB Blower Video Card (2-Way SLI) ($1612.98 @ Amazon) Case: Phanteks Evolv X ATX Mid Tower Case ($177.99 @ Amazon) Power Supply: Corsair Professional 1200 W 80+ Platinum Certified Fully Modular ATX Power Supply ($299.99 @ Amazon) Total: $4925.87 Prices include shipping, taxes, and discounts when available Generated by PCPartPicker 2020-02-06 15:53 EST-0500 CPU For multi-GPU usage I ideally have enough PCIe lanes. Theadripper 2nd Gen is cheap and supports enough lanes compared to the 3950x. While I can't upgrade to 3rd gen, I can always upgrade to the 2990WX which should last me another few years after the upgrade. MoBo The same argument as for the CPU is done for the motherboard. The motherboard should fit the sTR4 socket and contain enough PCIe lanes to fully utilise the GPUs. From the specs it says that 4 x PCIe 3.0 x16 (x16, x16/x16, x16/x8/x16, x16/x8/x16/x8) are supported. However, I found an infographic that apparently shows 3 x16 and one x8 GPU support. The slots should be far enough apart to fit all the big GPUs. The motherboard can also be extended to 128GB which will last for quite some time. Hardware Raid 1 support is also a nice bonus. GPUs Dual RTX 2080Ti is more than enough to get started. Thermal throttling is a big concern in multi-GPU builds so either I use blower cards or water cooled cards. The free slots can holds two more cards, either 2080Tis, or maybe even an RTX Titan at some point if I want the sweet VRAM. PSU While 1200W is too much for this build, if I extend it with up to two GPUs I will greatly appreciate this power. I also chose a multirail PSU. Storage & RAM An M.2 drive for the working data and two HDDs for cold storage that will be put into RAID 1. Maybe even with ZFS file system. (any opinions on ZFS?). 64GB ram is currently plenty. 3200 CAS 16 is apparently a sweet spot for RAM so I chose what seemed a good deal. Cooling A very important part is cooling. I plan an aio watercooler for the CPU and blower style GPUs. Blower GPUs rather than hybrid cooled GPUs because they are a bit cheaper and I might want to build a custom loop in the future to reduce the noise and cool slightly better. Case I think that the mid tower case is a bit tight for all the stuff but it should still work. This might be a pain point when I want to upgrade or do a custom water loop. The plan is to start with the above build and then add GPUs as the need arises for the local machine to be more powerful. After thermal issues or noise complaints from the neighbours I'll look into custom water cooling. Any comments or suggestions? This is my first build