CG Workstation – The “Apprentice”

Stepping up the game from both a n intel i5 and a AMD FX-8350 alike, the 4770K/3770K i7 “duo” is thet all-around processor for the demanding user. This build is aimed to fulfill the role of a mid-range 3D CAD workstation, capable of satisfying most casual users, amateur or professionals, with the ability to tweak and upgrade its performance overtime – either by adding GPUs, or overclocking the existing hardware.

To expand the capabilities of your hardware, quality components are key. Just like with everything in life though, and especially with electronics, past a certain performance point, you need to start investing much more for absolute performance as the value actually decreases.

This rig won’t be noticeably faster for casual modeling, photo editing, drafting etc tasks that a 3D Artist and/or designer will probably be engaged in comparison with the configurations featuring cheaper sub-components, like a i5 processor. The Hyper Threading (HT) capabilities of the i7 will be proven useful in workflows that favor heavily threaded tasks, like prolonged rendering sessions, video transcoding and developing large sets of RAW photos.

In this scenario the motherboard is a more sturdy choice that can allow for much more stable overclocks, while the 3x PCIe 16x slots are well layed out to accommodate up to 3x double slot GPUs for serious GPU acceleration in supported apps.

Note that despite the clear victory of Intel’s CPUs in single threaded applications, the FX-8350 remains competitive and even surpasses the Quad core i7s in pure VRay and Mental Ray performance.

  • CPU – LGA 1150: Intel Core i7-4770K. The newest addition to the Core i7 family, the 4770K is based on the Haswell architecture. Still 22nm like the 1 year old Ivy bridge, it doesn’t offer significant performance increase to justify upgrading for most people, but remains the fastest i7 quad core today. Clocked @ 3.5GHz stock speed, with a boost clock up to 3.7GHz, the 4770K is known to run hotter than both the older Sandy Bridge CPUs (2xxx codes) and Ivy Bridge  (3xxx), but that is not an issue outside those hopping for extreme overclocking. The K indicates an “unlocked” processor, where we are free to adjust the multiplier to whatever value we want easily through the BIOS or the Windows overclocking utilities of most decent motherboards with a couple of very easy and almost fool-proof steps.
  • CPU – LGA 1155: Intel Core i7-3770K. Not much to add here. The current LGA 1155 flagship from intel, offers 4C/8T with 3.4GHz base speed and up to 3.9GHz Turboboost. The 3rd Gen of intel core CPUs is based on the 22nm Ivy Bridge architecture, with excellent power efficiency and decent overclocking capabilities.
  • CPU Cooler: Depending on how you will treat your system, there are quite a few cooler options to go for:
  1. Cooler Master Hyper 212 Plus. If you feel the stock cooler is not enough, it is hard to beat the value of the 212…it is cheap, and very effective while nearly silent. Will keep your CPU happy even when mildly overclocked.
  2. Noctua ND-14. The mother of modern twin tower air-coolers, this huge chunk of metal with dual 140mm fans is here to keep your CPU cool even at high overclocks. Practically you cannot get better noise/performance unless you opt for a custom Water cooling kit. It does have it issues, being heavy and relatively challenging to install, but it does worth it. Costs around 2 times that of a 212, and it is clearly overkill for CPUs that won’t be overclocked.
  3. Corsair Hydro H100i. The current benchmark for AIO WC coolers, the H100i is not the best water cooling solution out there, but for the price it is hard to beat. The support from Corsair is also top notch, just like the warranty covering not just the unit, but also damaged components in case of a leak. It won’t impress versus a ND-14, but it is far prettier to look at, and much easier to install / remove. and won’t cause any interference with RAM dimms having large heatsinks. Costs around 3 times that of a 212, and it is clearly overkill for CPUs that won’t be overclocked. If you don’t plan to do so, but you still want to give it a go with a closed loop water cooler, a Hydro H60 will most likely suffice.
  • Motherboard – LGA 1150: Gigabyte GA-Z87X-UD3H. GIgabyte came up with the line having the least RMA requests in 2012, despite almost matching the leader in sales, Asus. The new, updated BIOS in the Z87 line, allows them to match the easy overclocking experience Asus users were spoiled to, making the Z87X-UD3H a pretty safe bet, just like the ASUS Z87-PLUS. Personally I prefer the black-blue color scheme the Gigabyte board offers much more than the gold/yellowish new scheme Asus came up with, but that’s subjective. Should you opt for overclocking your CPU, both should be able to do 4.5~4.7GHz as easily as tweaking your Vcore around 1.2V and setting the CPU multiplier for the desired speed ( 4x * 100MHz). The slot layout and connectivity options are nearly identical between the 2 models.
  • Motherboard – LGA 1155: ASUS P8Z77-V Pro . This offering from Asus pretty much sums up the qualities a high end motherboard should have without breaking the bank. While not offering any real performance updates over the VE/LK offerings from the same company, or any Z77 board from any manufacturer, its advantages come to the surface when you try to overclock. The board offers enough VRM phases for increased stability at higher Vcore settings, a great BIOS with a plethora of tweaking options, and a decent slot layout for up to 3x double slot GPUs. The Fan controls built into the motherboard are probably the best available, and the WiFi card is a welcome addition for anyone who doesn’t want or cannot use the built in intel Gbit NIC. USB 3.0 and Sata 6GB/s is natively supported just like with all Z77 boards.
  • RAM: G.SKILL Ares Series 16GB (2 x 8GB) 240-Pin SDRAM DDR3 1866.
    You can get 1x or 2x of these kits for 16GB and 32GB respectively.Planing ahead is wise. Most motherboards for intel i5/i7 and AMD FX CPUs have up to 4 slots for RAM. CPUs can support up to 32GBs. Even if you don’t see the need for 32GB of RAM, opt for 8GB sticks unless you want to go for a special speed etc. This will allow room for growth without parting out your initial investment (often the case with 4GB sticks). Prefer low profile heat-spreaders. Those offer little to no gains in reliability and stability, but might cause installation issues obstructing the use of large CPU heasinks, and even airflow to the CPU in tight situations. G.Skill Ares and Corsair Vengeance LP are good examples.
  • Graphics: This is a tough one. Depending on the direction you will move, this is the second most important component after the CPU, and in many ways the most important. I will list 3 options, trying to keep everyone happy:
  1. The AIO budget CUDA Renderer: EVGA GeForce GTX 770 SC 4GB. For CUDA optimized applications, and GPU accelerated renderers like iRay, VRay RT GPU* and Octane (among others), you have to opt for nVidia*. 2GB of Vram is enough buffer for most viewport** in anything but extreme applications and scenes, but GPU renderers can ask for more.  These cards can be combined for 4-way SLI, but compute tasks don’t care about SLI. As long as more than one compatible card with sufficient memory is detected, it will be utilized regardless of them being identical or physically bridged. 3D Viewports also couldn’t care less about SLI or Crossfire, as those are currently utilized only by gaming engines.  The 770 is a rebadged 680 with faster VRam, and the performance increase over the older 680s and 670s is not substantial. Given you can find them for a cheaper price, any 670/680 with 4GB Ram is a direct replacement with marginal loss in performance. A great example lately that would be the ZOTAC GTX 670 4GB, but supplies for 6xx cards in good pricing might not last for long.

  2. The budget conscious CADaddict: If all you do is 3D CAD and you don’t care about games or GPU accelerated rendering, then I would recommend considering and opting for a  nvidia Quadro K2000. Offers the performance of the older Quadro 4000, in a cheaper, smaller and cooler package. Low to mid range Quadros will do the occasional GPU accelerated task, but limited clocks and buffer sizes will be less impressive. Quadro’s don’t look the part next to the massive GTX cards, but trust me, especially in OpenGL viewports, a Quadro K2000 will be embarrassingly faster than the gaming cards. No surprises, even the Quadro 600 is faster! The K2000 is a valid replacement card even for the K4000 in suggestion #4.

  3. The All-out CUDA RendererEVGA GeForce GTX 770 SC 4GB, but rinse and repeat two or three times. Adding more cards, of course will speed up GPU computed views a lot. Viewports don’t care about SLI or multiple cards, so only the primary card having the monitor connected to it will carry the burden of accelerating your viewports. Unless you will be gaming “on the side” on your PC, make sure that VRay RT GPU, iRay etc renderers can accommodate your needs before you commit to buying more than one card. Also keep in mind that only the version of iRay that comes with 3DS Max 2013 supports the Kepler architecture and 6xx cards.
    The GTX 780 is faster, but since you can buy 3x 770 SC 4GBs for the price of 2x GTX 780 3GB boards , and since 2x GTX 770s (or even 2x 670s) are already outperforming a single GTX Titan, the 670 stops being the GPU of choice after your budget is comfortable thinking in the range of multiple Titan boards or a combination of 2-3 Titan boards and a Quadro/FireGL.

  4. The Hybrid Pro: nVidia Quadro K4000 + 2-3x GeForce GTX 770 SC 4GB. The best of 2 worlds…Some report issues with drivers, but from my experience with 6xx, 7xx and Titan GTX cards, I had no issues adding them as head-less accelerators to systems running with K2000, 2000 and 4000 Quadros as primary cards. Not a single glitch. Don’t expect the gaming power of the GTX cards on demand though. Switching monitor(s) to the GTX and re-installing drivers is a must if you want to game in a PC with both Quadro (or FirePro) and GTX cards co-existing. On the other hand, GPGPU programs and progressive renderers recognize all available compatible cards without issues.
  • SSD: Samsung 840 Pro 256GB. Probably the fastest consumer SATA drive at the moment, from the company with the best reliability record for the last couple of years. A 128GB drive is enough for most users wishing to add OS and basic apps, but 256GB will allow you for more flexibility + the ability to store most of your current projects without worrying too much about space. Make sure that you are opting for the 840 Pro series that is based on MLC NAND, instead of the “vanilla” 840 and 840 EVO that are based on TLC NAND and will supposingly have a shorter lifespan. The price difference is not important enough to risk a drive failing prematurely (all SSDs of current technology eventually will). Note: SSDs are offering considerable boost in loading times for all apps and the overal responsiveness of the system. Still the performance benefits for most CG applications as far as rendering speed and viewport performance are negligible or non-existant, with the slight exception of video editing with multiple streams of HD video. A smaller, or no SSD is the obvious choice for the budget conscious builder that doesn’t want to sacrifice real performance.
  • HDD: Western Digital Caviar Black 2 TB. Most >1TB 7200rpm drives are fast, but the WD Black line is at the moment the only one offering 5 years warranty, while competitors (and WD Green/Blues) give 2-3years. I would get the extra coverage over 5% performance any day. Combining two of the above drives in RAID 1 (mirroring) mode, will allow for the much safer storage of critical data in your system.
  • Case: Antec P280 Black Midi Case. There is a huge selection of cases available out there, offering a variety of features and aesthetics. The P280 is a simple and elegant design that works, offers some sound-dampening lining, has decent cable management options, can fit most large air coolers and closed loop AIO water coolers up th 240mm like the Corsair H100 without any modifications.
    More importantly, it was chosen for the 9 expansion slots, that allow for more flexibility when planning to use 3x double slot GPUs, as the different arrangments in the PCIe 16x slots in many motherboards make it hard with most 7 or 8 slot cases – usually for the cards being in the far bottom slot. The P280 is one of the few 9-slot Midi ATX cases available, and it is pretty good value. Feel free to substitute with any unit that suits your aesthetics, given it will accommodate your equipment properly.
  • PSU: Corsair AX860i 80+ Platinum or Cooler Master V850 80+ Gold
    The sole reason for suggesting such a high-power unit, is for the manufacturer’s provision for 6x PCIe (6+2) auxilary power connectors, capable of feeding 3x high power GPU cards for computation. If you don’t plan on going this way, do not bother over-spending. It is important to mention that most systems with one CPU/one GPU, even with overclocked components want be drawing more than 350-400W of power from the wall. People get over-zealous with their PSUs, opting for high capacity units that most likely will be greatly underutilized. It is not wise to get a PSU that you know will be stressed to more than 75% of its capacity for prolonged time, so I would never advocate on getting a 450W PSU even for a single GPU workstation, though a 550W would more likely be more than enough and a 600-650W would probably be sufficient for adding a second GPUcard. Most likely the next generation of GPUs will be more power efficient, so you will end up using even less power should you upgrade in a couple of years. The 80+ Gold and above certifications, usually are limited to high quality units, so opting for one nearly ensures longevity ontop of measurable energy savings over time. If you already own a 80+ PSU that is relatively new, depends on the country, electricity rates and usage patterns whether you will ever break even upgrading to a better 80+ rating, but reducing waste is always good even if you end up $5 in the whole over the lifetime of a product.
  • Optical Drive: Optical Drives (ODs) are rarely utilized the last few years. Broadband connections, cloud storage and affordable flash drives are replacing them. For most ODs are limited to installing new software and OS. A Lite-On IHAS124-04 or Asus DRW-24B3ST would do the trick.

  • Operating System: Windows 7 Professional SP1 64bit (OEM) Windows OS is guaranteeing compatibility with most rendering packages, including 3DS Max. If your distributed rendering will be focused in Maya, you could actually get away with Linux as client packages for VRay are available.
    Windows 7 Home editions are limited to 16GB of RAM. Professional and Ultimate versions allow up to 128GB, and offer some additional networking/remote access features that are desirable.

Disclaimer: some of the links contained within this post and resource pages have my referral ID, leading to a small commission for each actual sale  that originates directly from my material. This doesn’t mean I won’t strive to provide you with objective information and honest opinions that will potentially help you on educated decisions. If you like the information in this site, please support it by using the links below when you decide to buy.

29 thoughts on “CG Workstation – The “Apprentice”

  1. Hi, Thank you for the post. It is very helpful! I have a very quick question about GPU rendering (VrayRT and Cycles). Primarily I am going to make architectural stills and small fly through animation 15 – 30 sec Tesla/Quadro thing is way beyond my capabilities, so I am choosing between : – one gtx titan (potential upgrade to two titans in one year) – two gtx 680 (EVGA 4GB edition) – three gtx 670 4Gb Questions: – is it possible to make animations with VrayRT (15 to 30 sec) – which of this Graphics cards will perform better I know that Titan has 6Gb of Ram I also asking because I want be able to update my PC for at least 2 years. My budget is 1000-1200$ Thanks

    • Yes, you can render small (or large) animations with VRay RT. It has been the case for many studios over the last few years.
      The titan is no “magic” card. It does have much better double precision calculation performance (as far as I’ve seen in reviews) but that doesn’t matter for VRay RT. For its biggest part it is still based on the same CUDA architecture as all GTX Kepler cards. Lets see how that aggregates:

      GTX Titan – 6GB@GPU: 2688 cores / 876MHz (boost) = ~2355 GHz ~ 2.35GHz/$ spent

      GTX 690 – 2GB@GPU: 3072 cores / 1019MHz (boost) = ~3130 GHz ~ 3.13GHz/$ spent

      GTX 680 – 4GB@GPU: 1536 cores / 1058MHz (boost) = ~1625 GHz ~ 3.07GHz/$ spent

      GTX 670 SC- 4GB@GPU: 1344 cores / 1046MHz (boost) = ~1406 GHz ~ 3.20GHz/$ spent

      I hope you see the picture around me picking the 670 SC (the cheaper 4GB 670 anyways where I’m at) as the base of a CUDA renderer.
      With the potential of the 690 being ram limited in complex scenes and 2GB per GPU, the Titan is still the best choice if you want to setup the fastest machine with 2 or 3 cards and you are not budget limited, but 2x GTX 670 will be faster (and ofc 2x 680s also). Assume you were able to find Titans which get out of stock everywhere due to nut gamers, hardware reviewers etc, you can still buy 3x GTX 670 4GB for 30% more than a single GTX Titan, and get ~ 180% the performance (or 90% the performance of 2x Titans).
      If you get the 3770K as a base of your PC, I am sorry but you won’t be able to upgrade once you hit 3x cards (4x with certain motherboards, like the Asus Z77 V Extreme. The socket won’t be used in the next generation of Haswell i7s, so only s2011 has newer processors coming its way.

      If you have $1200 for GPUs, I would definitely go the 670 way, squeezing in a bit more to get 3x, unless I was PSU limited, where I would opt for 2x 670s and wait. I don’t think the 680 is enough faster to worth nearly $100 per card.

  2. Dimitris, Thank you very much for your reply. I have read a lot of review about all GPUs out there sometimes it is getting confusing. But after all the reading and your post I think I will go 3x GTX 670 way. One thing is worrying me is : Is it worse investing to old 670 with 4Gb when in one year I can get more advanced GTX kepler gpu with more memory? VrayRT 2.4 have added kepler support already. Here are my build spec: Motherboard – Asus P9X79-WS Processor – Intel Core i7-3930K Processor (12M Cache, up to 3.80 GHz) RAM 16GB G Skill Ripjaws Z 16GB DDR3 PC17000 2133MHz Kit Power supplier – Seasonic SS-1250XM X-Series 1250W Modular Power Supply (PSU) 80+ GOLD And the last one: Will it be possible to render this scene with 4Gb of ram: http://www.archithings.com/som-wins-master-plan-competition-for-beijing-bohai-innovation-city/2012/09/21/som-wins-master-plan-competition-for-beijing-bohai-innovation-city-02 Thanks for help and sorry for bother you. Cheers

    • No idea on the cityscape – it clearly depends on the tree models and how those are proxied / how detailed the original models are etc. The buildings ofc are a “joke” – super simple.

      As far as the hardware progressing, there is nothing we can do…if we think like that, we either are in an infinite regression cycle where nothing gets “done”/bought etc, or we draw the line and say, “there, it is good enough for the budget, we will make it work with what we have today”.

      Amazing CG art is created with far less what’s available today, did not stop them, should not stop us. We just have to be more creative instead of “brute forcing” it in 😉

  3. Hi Dimitri, Im thinking about buying this configuration but I would like to choose an i7 4770k instead of the 3770k, what motherboard do you recommend me for that CPU? Should I change more components of this configuration for the 4770k? Thank you! 😉

    • Updated the config after your comment. I believe that the Gigabyte GA-Z87X-UD3H is a great “mid-range” board that offers everything you or most users might need.
      The only real benefit for a “more expensive” board like the
      Gigabyte Z87X-UD5H or an
      Asus Z87 Hero , would be the desire for pushing your CPU to the limit. I don’t know if an extra 100MHz or so worth it when you can already hit a 1+GHz overclock with boards 2/3 the price.

  4. Thank you very much Dimitris, coming from and old Quad Core CPU I preffer to choose the 4770k instead the 3770k. I think that Gigabyte motherboard is good enough for me. Cheers!

  5. Hello Dimitris, Compliments on the constructive and clear elaborations on possible configurations, that’s a talent !! I’m new to the world of DIY pc systems and getting more and more hooked on the virus called visualization. I work with Autocad 2010-14 / Sketchup 2013 + Vray ( RT ) / Adobe PS – Illustrator – Indesign. In de near future Revit / 3ds max / Vectorworks will be tackeld. I’d like to take full potential of Vray RT, because I think this can save me alot of time due to it’s direct feedback aspose to setting up / render / reset / render etc, etc. I thought I’d made up my mind with the following setup; Budget : € 1.300 ( inside components ) CPU : i7-4770k ( i7-4930k, think this is an overkill for my first overclock CPU ) Mobo : Asus Z87 Pro ( like the Btooth + Wifi ) GPU : MSI Geforce GTX 760 2Gb ( Vray RT ) Q 1 ; Roaming the net I get the impression that Intel CPU’s set the pace for Render workstations. You also point out the AMD FX-8350 as a CPU to take into account. I’m puzzeld and seems like a toss. But that’s of cource not the way to go. Could you point out, with my needs above, which one cuts the edge and why (or even another one, just to make me puzzel more.. haha ) ? Q 2 ; Will the GTX 760 2GB be adequatly enough for my purpose with Vray RT? Or go for the 4GB version at + €100. Cheers

    • Jakob, What you list is a solid setup. The FX-8350 is a great CPU for anyone on a tight budget – especially if he/she will be rendering and not just modeling vs. its real price range competitor = the i5 line. The i7s of the last generations are usually ending up ontop of the FX-83xx for both single threaded and multi-threaded tasks. You would not be disappointed with an 8350, but a 4770K is an all-around faster CPU, and if you can afford it, by all means do.
      The 4930K is quite a bit off your price range for sure. A good mobo and that CPU would deplete the better part of your € 1,300 budget.
      GPU: The newer adaptive degradation engines in 3DS Max 2014 / Revit 2014 are – at last – allowing us to work comfortably with GTX cards. Certain aspects of viewports are still driver / CPU bound, so it is not perfect, but nothing is.

      Sketchup is doing is own thing…had that kind of adaptive degradation (for grouped geometry) long before Autodesk tried it, and should work fine with most modern GPUs – tho being OpenGL will favor certain Radeons a bit more than GTX Cards. The big but: VRay RT. You will need a GTX. Did not try the VRay 2.0 for Sketchup yet, but if it is anything like the 3DS version, the OpenCL mode works horibly with AMD cards. Chaosgroup is blaming AMD drivers and whatnot, but I don’t trust them as AMD manages to work wonders with Adobe and other OpenCL accelerated apps. That said, the 760 is an fine mid-range card to use. 2 or 4GB is purely decided on texture sizes and model complexity used (dont’ know if VRay RT for Sketchup shrinks textures and to which level). The price difference for the 4GB is indeed steep, especially if you factor in the fact that 256bit cards cannot really access 4GB of RAM simultaneously for them to count…works for VRay tho…try to find an offer on a phased out GTX 670 if you can…its slightly faster than the 760, and I’ve seen some amazing offers on them few weeks ago, as shops were pushing stock out.


      Vectorworks / Maya / Rhino etc OpenGL based programs work “ok” with GTX, a tad more OK with Radeons and can give you trouble with complicated models, but with proper workflow and some patience it should work just fine with any modern card. Don’t forget a good cooler since you’ve mentioned overclocking…a H80i would do fine, fits in most cases. A Swiftech H220 or H100i is ofc better, but do some research on the case you want to buy and use with those to make sure you can fit them properly.

  6. Dimitirs, thanks for your reply. Toggeling with my budget I’m also thinking about an Xeon E3-1230 V3. Budget wise between the AMD FX-8350 and i7-4770(k). I’m assuming to have the advantage of Intel’s betere single and multi tasks performance. Q 1. What’s you thought on this ? Otherwise I’m thinking just to go for the i7-4770(k). Attempting to get my head around GPU cards, I think you are saying that software ( Vray RT ; possibility of shrinking textures ) accually dictates if a GPU card can cope with the 2GB or 4GB version !? Your mentioning that , ” 256 bit cards cannot really access 4GB of RAM simultaneously for them to count “. Q 2. Does this mean 384 bit do, or what is it what makes them accually do? You also adviced to look at the GTX 670 being faster than the GTX 760. Q 3. Is that because of the more CUDA’s of the GTX 670 ? Thanks for all your explanations and will be looking at those other important compontes to take into account. Cheers

    • A1) The E3-1230 V3 is not a bad option. It is 200MHz slower than the 4770K, lacks the built-in graphics, but it is also some $80 cheaper. Ofcourse you forfeit any chance of overclocking with it, but it remains as fast as any haswell @ relative speeds.
      A2) 256bit cards – like all 6xx/7xx but the 780 and Titan – are proven not to gain any performance going to 4GB over 2GB, as the memory controller cannot access 4GB of RAM effectively. That is notable in gaming, where things need to be accessed pretty fast. You won’t get more fps in games and viewport going 4GB over 2GB in most scenarios.
      In GPU rendering engines, the extra RAM does help as far as rendering speed goes, but does help you for loading and starting GPU renderings for larger models and/or large textures. Remember, textures in the GPU are “decompressed”, so a 1024×1024 texture would take up 3MB or space, a 2048×2048 would be 12MB and large HDMI panoramas would be in the hundreds of MBs. Engines like VRay RT, would initially automatically resize all the bitmaps to 512p before initiating a rendering, to save up space in the CPU. Now this is user selectable.
      A3) Ram is still used in 64bit packages/chips. A 384bit bus allows more data to be pulled or pushed through it, by accessing 6 chips in parallel, instead of 4x for the 256bit cards. Depending on the application and the GPU core (how memory starved it is), 384bit might help. Kepler architecture is offering pretty scalable performance. Real life data (rendering times) show that each Titan’s CUDA core performs pretty much as well as a 760’s CUDA core in GPU rendering / compute. The higher ram speed doesn’t really help such rendering engines. Does a lot for other compute applications (like F@H for example), but it is not a universal performance boost, nor a requirement to max out performance.
      Yes, the “aggregate” speed of the 670’s cores (MHz per core * # of cores) is a tad higher than that of the 760, thus it performs a tad better all around.

  7. Dimitris, Thanks for your thorough explanation. Still got a lot to learn about hardware. Well got to start somewhere… Dit some homework and found this ; http://www.playtool.com/pages/vramwidth/width.html , to understand the basic principals of the VRAM on the GPU card. Really helped to understand your input. Will continue on my quest for futher componets. Cheers.

  8. hi thank u for the post. i use these software: 3dsMax 2013 (Modeling,Rigging,ParticleFlow,MassFx,Rendering(MentalRay)) PhotoShop CS6 AfterEffect CS6 i have about 2300$ what system recommended for me? Thanks 🙂 … one last question can i buy 2case with Fx-8350 and get more speed than one case with i7 4930k?

    • For this budget a 4930K based system would be the preferred choice. 2* FX-8350 might be as fast or a tad faster in many cases, but that would be in pure rendering (heavily multi-threaded tasks). 3DS Max and Adobe CC are still heavily reliant on absolute single thread performance too, where intel CPUs have clearly the upper hand.
      Newly developed applications with proper multi-threaded routines (like AA game titles) let the FX-octacores shine far more than the average digital content creation program.

      • thanks 🙂 one question what do u think about these list? which one is better? first: ASUS P9X79 Intel Core i7-4930K ThermalRight Silver Arrow SB-E Extreme Cooler Patriot Viper3 Blue Sapphire 32GB 8GBx4 2133Mhz Sapphire R9 290 4GB Seagate Barracuda 2TB 64MB 7200Rpm Plextor M5 Pro 128GB SATA3 … second: gigabite ga-x79s-up5 wifi rev(1.1) Intel Core i7-4930K water cooling :ELC240 AiO liquid Gskill tridentX 2*8GB=16GB DDR3 2400MHz GTX 770 OC WF 3X 2GB GDDR5 seagate SATA lll 64MB constellation CS 2TB a-data XPG SX900 128GB … third: (your suggestion 🙂 ) …………. thank u 🙂

          • thanks 🙂 one last question which one is better for GPU rendering? gtx 780 ti or radeon r9 280x ? http://compubench.com/result.jsp

            • Depends on the GPU renderer. If you would be using a renderer properly written in OpenCL, AMD cards would have a clear advantage. In most OpenCL benchmarks, and rendering engines like the Lux Renderer (including the Luxmark benchmark), even a 7950 is much faster than a GTX Titan or 780Ti.

              For CUDA based renderers though, like Vray RT or Octane, the nVidia cards are the only way to go. Don’t be fooled by VRay RT GPU having a OpenCL mode: it is actually directly recompiled from the CUDA version (i.e. running CUDA in “emulation” through OpenCL), and AMD cards either fail to compile it or run VERY slow.
              The 780Ti is currently the fastest consumer card for CUDA accelerated apps, with GTX Titan following closely. The Titan has 6GB of buffer, giving it an advantage for complicated scenes over the 780 cards that are limited to 3GB.

              If I would not get a Titan, I would stick with the cheaper, “vanilla” 780. It is not that slower than either 780Ti or Titan, but considerably cheaper than both. It is limited to 3GB, but so is the 780Ti.
              If 3GB are enough, and you need a faster card than the 780, then maybe going 2x 780 will be a preferable and truly faster choice than spending $250 or more to get the 780Ti.

              nVidia is rumored to release the 790 (dual GK110 card with either 2x3GB or 2x5GB RAM), and rumors are also heard about a “GTX Titan Black Edition”, that will probably feature a full GK110 with 2880 cores (same as the 780Ti and K6000) and probably still 6GB of RAM. If anything, those releases might give you more options, but also push the prices of the existing GK110 cards down.

              • Mental Ray and VRay Adv remain exclusively CPU based, thus neither CUDA or OpenCL.
                The iRay renderer that comes with 3DS and is GPU accelerated, just like the VRay RT GPU renderer – a separate engines.
                iRay is developed by nVidia, so unlikely to be OpenCL soon. Vray RT GPU will probably get better OpenCL support in V3, but I haven’t been in the beta program to know how much better works than the current version.

  9. hi Dimitris thanks for your answers 🙂 one more question 😀 which method is better for animation rendering? Gpu Or Cpu? i really confused 🙁

    • Not an easy answer.
      Depends on the animation and the VRay features you want used in them. VRay RT GPU doesn’t support all the features of VRay Adv ( CPU only). It is not the same engine. IF, what you do is “covered” with what VRay RT GPU supports, and we are talking an investment in hardware above a minimum level, then GPU rendering is truly faster per $ invested.
      That doesn’t mean that ANY workstation with a $250-300 GTX will render always and everything faster than a $300 i7 CPU.

      But If you were to go deep into your pockets, and start comparing what – say – a $6000 dual Xeon investment, to a $3000 3xGTX Titan investment and compare them in rendering simple animations – like not too complicated architectural walkthroughs for example – the GTXs will probably render frames considerably faster.

      This again presupposes you are molding your workflow within what VRay RT GPU can do, and not the other way around. Vray Adv is a more flexible solution, but some of the extra features do add to the compute time.

  10. I followed your great post advice and I’m putting together a system. However PC Partpicker is showing 2 incompatibility issues: 1.- “Intel Core i7-4770K 3.5GHz Quad-Core Processor and Asus P8Z77-V PRO ATX LGA1155 Motherboard are not compatible” 2.- “Antec P280 ATX Mid Tower Case and Corsair H100i 77.0 CFM Liquid CPU Cooler compatibility could not be evaluated.” Should I worry about that ? Is the CPU+MB issue a Bios setting problem? If so how is it to be set? Please advise. Thank you !

    • *PC Partpicker is right: the 4770K is a socket 1150 CPU, and the P8Z77-V PRO is a socket 1155 Motherboard. No BIOS can fix that discrepancy. If you check the article, there were suggestions for both s1150 and s1155 motherboards.
      If you don’t plan on overclocking, any H87 or Z87 motherboard from ASUS or Gigabyte will serve you well. No need to over-spend with “Pro” or “Workstation” models. I recently bought a Z87 motherboard for the 4770K bench PC I was building, and I ended up with the Gigabyte GA-Z87X-UD3H, which seems preferable as far as VRM quality to ASUS offerings before you hit the $200 bracket with a Z87-Hero.

      *The P280 does fit the H100 just fine. You can google for “Antec P280 h100 ” and go to “images” to see the plentiful clearances on complete systems. There was no “magic” in the P280, you can get many midi towers with the same functionality. It was just a serious looking case that would fit most CLC coolers (like the H100) and had enough pci slots for triple or quad dual width GPUs. If you don’t plan on multi-GPU setups, you can get a smaller / cheaper case. Even a mATX with the corresponding mATX motherboard. There will be no performance degradation between mATX and full-ATX motherboards that use the same chipset.

      • Hi again, sorry I saw your post after I posted mine … I needed to refresh. Expanding a bit on your suggestions, : So if I understand correctly, there’s no way to overclock with a “non-pro’ MB model? I wonder whether in the future I might get the OC fever and I decide to try it … 🙂 With regards to have multiple GPU’s is definitely not likely, so I like the idea of having a smaller tower. So the “Antec P280 ATX Mid Tower Case should be fine but when you say “smaller” like any mATX … I thought hat mATX stands for mid case but I guess it’s for “mini Case? Correct? Also I’m a bit unsure about choosing the right mATX mobo equivalent to the Gigabyte GA-Z87X-UD3H . How do I search for tthat? Thanks so much for your patience. I’m fairly new to that stuff as you can tell.

        • *Almost all Z87 motherboards from most manufacturers (Asus, Gigabyte, MSI, Asrock) allow you to overclock just fine. H8x and B8x series chipsets don’t have overclocking provisions built in or allowed in their BIOSes.
          What separates the various motherboards – since the chipset and minimum specs are all sourced/over-viewed by intel directly – the premium motherboards add additional USB slots, above average all the way to “overkill” VRM circuity (the voltage regulators that ensure proper power delivery to the CPU when overclocking), additional controllers for RAID, LAN, WiFi, PCIe lanes and/or higher quality onboard sound cards.
          The Asus “Pro” mobos are mid-range. Their ROG line (Hero, Gene, Formula, Rampage) are the high-end overclocking+gaming line, and the WS are the “workstation” line, with each chipset having potentially a dozen or more models from each manufacturer that follows its own naming pattern. Usually when you buy a Z77/Z87 board, you are ensured to be within 2-3% performance of the best, regardless of origin for base clocks. The benefits come from the peripherals and maybe how easy or high you can OC. So “minimum” performance and quality is pretty much guaranteed.


          * Motherboards come in 3-4 main sizes. ATX/E-ATX that have usually up to 6 PCIe slots, micro-ATX (mATX) that has up to 4 PCIe slots, and mini-ITX that has one PCIe slot. Since powerful GPUs are usually 2-slots wide, you can go up to 4* in ATX/E-ATX, up to 2* in mATX and ofc 1* in ITX.

          * Midi Cases are accepting full-size ATX motherboards or some of them even bigger motherboards. Their “medium” size is by comparison to full-ATX cases that are those “barely-fitting-under-the-desk” sized towers.

          * A mid-range (price wise) Z87 mATX I would recommend is the ASUS Z87M-PLUS (~$130), and if you want to go for the “better” yet not extreme, would probably be the ASUS GRYPHON Z87 (~$160). Gigabyte doesn’t have very decent mid-range mATX boards, despite having probably the best ATX ones. Both Gigabyte and Asus have great overclocking mATX boards in the $200 range, but I think that’s heading towards the “best-off” and not the “best-value-for-money”.

          *There are a few decent mATX cases around with the Corsair 350D being probably one of my favorites ($70~75), but the Cooler Master N200 also very good in its price bracket (<$50).

          • Dimitris, I really want to thank you for your expert support and spending your valuable time helping those of us without hardware experience in today’s huge selection range. Here is the ‘final’ (for now) build: CPU: Intel Core i7-4770K. Cooler: Corsair H60. MB: Asus Z87M-Plus Micro ATX. RAM: G.Skill Ares Series 16GB (2 x 8GB). SSD: Samsung 840 Pro Series 256GB. HDD: Seagate Barracuda 2TB 3.5″ GPU: Gigabyte GeForce GTX 770 4GB (**) Case: Cooler Master N200 MicroATX. PSU: Corsair Professional 650W 80+ Gold. Wireless Network Adapter: TP-Link. TL-WDN4800 802.11a/b/g/n PCI-Express x1 Wi-Fi Adapter Total: CA$1835.13 (incl tax &shipping. Not bad at all) (**) There are a 5 GTX770’s 4gb cards at EVGA website http://www.evga.com/Products/ProductList.aspx?type=0&family=GeForce+700+Series+Family&chipset=GTX+770 I don’t see the ‘770 SC’ that you suggest. I picked the lowest priced 770 at Newegg. http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&DEPA=0&Order=BESTMATCH&Description=GV-N770OC-4GD&N=-1&isNodeId=1 Anything wrong with that? It seems to be a huge price difference between models/brands. I should mention that I’m building this rig for Rhino+VrayRT/Lumion/C4D animation (not worried about 3ds Max viewport efficiency as I use Max very rarely. Not particularly huge models (1.5 M polys at the most). I would appreciate your final blessing … I’m ready to order ! Huge thanks.

Leave a Reply to Roman Cancel reply

Your email address will not be published. Required fields are marked *