How to play hd movies on pc?

What would be the specs of a PC that could be used to watch HD movies, play high graphics games, and use advanced software like Matlab, VisualStudio, and AutoCad?

• Answer:

  Understanding that the choice to game will impair your ability in AutoCad and other professional graphics applications there are some options, but none of them are good. It cannot be underscored enough that AutoDesk has repeatedly released performance data that demonstrates their software runs better on Quadro / FirePro cards than it does on GeForce / Radeon cards. You can throw more cuda cores, more v-ram and a higher clock at the application and get worse performance than you would from a card with the right chipset. At the same time, games don't care about how many cores you have on your CPU, there is very little to be gained from gaming on an i7 instead of an i5, and there is benchmark data to suggest that when using the same GPU, in many games there is no user noticeable difference between a 2 core non-hyperthreaded Pentium AE and an i7 with 4 hyperthreaded cores. What matters to games is Clock Rate, not Core Count. So what about MatLab? MatLab doesn't make use of Hyperthreaded 'virtual' cores, but it does use multiple cores.  In this case, you may be better off with a Xeon processor for more physical cores (say 12 cores on single CPU) or an AMD FX Series that uses more physical cores (that's debateable, I haven't seen data for MatLab on AMD hardware.) Ultimately, in building a PC to play games and run computationally intense applications, you're going to have to choose. What's more important to you? If I had to split the middle ground and opt for dual purpose here's what I'd do (Note: With X99 mere weeks away, I'm going to give an answer that discusses both X99 and X79) First, if I was building today, it would be an X79 motherboard with an i7 4930K. Why? Balance. X79 will get you into 40 PCI-e lanes for GPUs (talk about those later) and it will also let you leverage a 4930K for six physical cores (12 threads that may or may not be useable.) The processor is going to pump out power for AutoCad (a little less so for MatLab) and will out perform a 4790K or most AMD chips handily. At the same time, it refrains from wandering into the gaming unfriendly lands of Xeons which generally clock lower and are most viable in dual processor configurations anyway. For X99 I'd be looking at a 5960X to feed MatLab an additional 2 physical cores. The hit in clock won't be horrible for game play and you can always invest in cooling to recoup some of the loss. GPU wise, I can't find any data to support AutoCad making better use of Cuda or OpenCL, so take your pick. Given that we're already taking a performance hit for using a gaming card instead of a workstation card, I'd opt for dual cards in either SLI or Crossfire. Dual R9 290s would be my go to for budget, but there's nothing wrong with throwing some Titans at it if you have the budget to do so. Let your gaming preference and your budget dictate your brand choice. Given that we're looking at a performance engineering piece of hardware and that the folks at AutoDesk tend to recommend a lot of ram if you're doing 3D visualizations, I'd opt for 32-64GB of RAM. With x99, that's going to be expensive as X99 will be the first chipset to feature DDR4, but there's no reason to let system resources bottleneck your work flow. AutoDesk recommends 16GB as a minimum for 3D visualization workflows. For most of this, I'd opt for a reasonable size SSD: 512GB, 1TB if the budget allows, but the rest of the data can safely be put on spinning platters, as many as you need. I'd take the time to invest in cooling. A closed loop water block would be a good place to start, and given that AutoCad doesn't make use of GPU rendering, that may be as far as you need to go. If you're doing extended gaming sessions, expanding cooling to an open loop to get liquid cooling on the GPUs would be worth while, but otherwise, probably just adds expense to the build. Okay, I know that's a huge block of data, but I hope it illustrates that the hardware needs of professional software and the hardware needs of gaming software are different and while there is overlap, if you want to go far in either direction, you start working against your ability to go the other direction.