[Dave Haynie]

That’s certainly a beast of a laptop. I have a new one with basically the same processor, a lower-end nVidia GPU, 16GB, PCIe-based SSD (512GB), and the best display on any laptop (4K quantum dot), but only 15.6″. Unfortunately, they seem to have pretty much abandoned 17″ screens. And I wanted to actually have battery life — I have desktops at home and work, so I’m not trying to entirely replace a desktop with a laptop. Because you really can’t, not for anything that needs lots of CPU or lots of screen.

Your GPU is a really fast mobile GPU, don’t worry about that. It’s nVidia’s latest architecture, Maxwell, and with 1280 rendering pipelines, it’s no slouch. In fact, that’s twice the size of mine, and the bus is 192-bit, versus 128-bit on my system.

As other have mentioned, you’re not going to get Main Concept AVC fast rendering with that chip. Maxwell architecture is replacing Kepler, Kepler replaced Fermi — which is the last architecture Main Concept supported. Now, keep in mind that whey they did here was pretty evil. The whole point of the CUDA or OpenCL interfaces is that the GPGPU computing doesn’t count on the specific GPU.

But Main Concept hard-wired these in, basically to give them leverage over companies like Sony. Then they were bought by DivX, then Rovi, and basically ignored the product. At least based on what they’re showing on their web site, Sony has the latest version — they just pretty much abandoned new GPU support.

The rest of Vegas, like some plug-ins, compositing, etc. will benefit from the GPU. You need to use the latest OpenCL drivers from nVidia, and unfortunately, they haven’t been as good with OpenCL as they are with their proprietary CUDA. And of course, you have to enable the use of OpenCL in the “Video” tab of the preferences dialog.

Just looking at basic benchmarks, the 850M in my laptop will beat my desktop AMD HD6970 on maybe half of the published benchmarks.. the HD6970 may be the fastest GPU for Main Concept rendering. But it’s a 2010 card… things evolve. So that’s one of the fastest GPUs you can get in a laptop without going to SLI. Try using the “Red Car” benchmark from Sony, that’ll definitely take advantage of GPU compositing, even if the AVC rendering is the same. Just rendering to Main Concept from any single video source, you’re counting on the AVC acceleration only. Which you aren’t going to get.

-Dave

[Dave Haynie]

Hmmm… I actually have that very same mouse, the G500. I suppose I chose it because I trust Logitech, and as well, it’s one of the few gaming mice I found that didn’t look like it might belong to a 15 year old kid.

-Dave

[Dave Haynie]

I never formally learned to type. But between typing all the time, and keytops wearing out, I’m pretty much touch-typing at this point. You know, you do something for 40 years or so, you may eventually get the hang of it.

I use the ShuttlePro on the left, too. That doesn’t get enough abuse to be in need of repair any time soon. I usually use a gaming mouse. They cost a bit more, but they’re rugged, mouse and cable, high resolution (I’m working across three screens), and they include a set of weights, so you can make the mouse heavier if you like, which is kind of a good thing.

For the keyboard, I just want something with good action. Too many of them feel like the keys are supported by gumdrops.

-Dave

[Dave Haynie]

If you’re a Vegas user, Windows 7 is a good guess anyway. But this was the only driver I had — if it didn’t work, you weren’t going to be any worse off.

I got a ShuttlePRO V2, a stand-alone shuttle controller, and a fairly ordinary keyboard (Lenovo) with its own USB jacks for shuttle and mouse. More expensive the first time around, but it’s a better keyboard, and when it does die, I’m pretty sure the shuttle and mouse will not need to be replaced at the same time. Modularity!

-Dave

[Dave Haynie]

I used to have this keyboard. A little disappointing that, like so many, it just had painted-on keycaps that faded. And it didn’t last… but I’m not very nice to my keyboards.

Anyway, I just dropped the driver here: https://www.frogpondmedia.com/dump/JSM_SetupPC_3.0_Final.exe

-Dave

[Dave Haynie]

I’d really like to try a new GPU, too… probably wait a little. Because there’s this whole landslide effect. Right now, my AMD HD6970 is pretty much the go-to card for MainConcept AVC, but it’s clearly not a performance leader anymore for GPGPU stuff. My new laptop has a, well, laptop-class nVidia, the new Maxwell architecture, and it’s beating the desktop on about half of the GPU benchmarks. So a proper modern desktop card ought to be pretty nice, indeed.

But the HD6970 is limited to 2560×1600 displays, which explains the two 2560×1440 displays on my desk (with a 1920×1200 as an adjunct). But were I to get a new CPU, then it would only make it easier to replace these with a pair of 4K monitors. And now you’re starting to talk some real money. And then, once the 4K monitors are up, I couldn’t but help notice that there’s this Panasonic GH4 camera that I have a bunch of lenses for (mostly Olympus m43 system as a walk-arounder when the Canon’s not needed) that shoots 4K. But most of those are primes, and I’d probably want some good zooms. And once I have all that stuff here, I’m going to start looking at the media room and that there’s “only” a 70″ full HD television there. Not even all that old. And once editing 4K on a new GPU, I’d certainly notice that maybe six cores isn’t enough.

Bad thoughts, all that stuff… maybe it’s the new work-paid laptop.

-Dave

[Dave Haynie]

[John Rofrano] “[Scott Francis] “I really appreciate all your input over the years and this is very informative. ”
You’re welcome. Hey, I was the one who always said that Mac’s were overpriced before I ever owned one. Then when I was looking for a new computer, I priced the specs of the Mac Pro against HP Z Series and BOXX Computers and the Mac Pro was over $1000 cheaper for the same parts. That’s when I realized that at the high end they were very competitively priced. “

Yup. Apple’s unique in that they maintain their margins even at lower priced gear. So they’re pretty expensive when you’re looking at entry level systems, even though they’re doing at least as much volume as most any other PC company, and more than most. Of course, when you look at guys like HP and all, lucky to make a 5% margin on an entry-level PC (and many of the Asian brands not even going that high), you do have to wonder how they make a living.

On the high end, it’s a very different dynamic. There just aren’t that many people who need a higher-end PC anymore, so the volumes are much lower. Companies like Boxx specialize in just this kind of PC, so they’re actually dependent on higher margins than Apple to survive. They make nice stuff, but it’s quite expensive. I needed to finally get a laptop for the “Day Job”… electronics CAD, which of course doesn’t really have a laptop market, but works pretty well with either video or mechanical-CAD oriented systems, though mechanical-CAD is generally way overpriced, thanks to the pro 3D graphics cards. Box with the basic configuration I was looking at ran about $1000 more than the one I wound up with, an Asus NX500JK-XH72T (which, incidentally, has the best screen I’ve seen on any device… a new quantum dot technology, rather than conventional LCD. An the PCIe SSD is about twice the speed of my desktop’s SSD! Not that I’d really recommend a laptop for video work, but this might actually manage HD editing ok).

I took apart some Macs back in the Commodore days. I didn’t like the electronics design — they did way too many things in software that should have been done in hardware, but of course, they didn’t multitask in those days. But the mechanical designs were very nice. Back in those days, when the FCC wasn’t looking too hard, they had velco-on tops for the Mac IIs, and a power supply (made by Sony) that plugged directly into the main board. All other things went to the main board as well, none of this maze of wires you used to find in a traditional PC (including our Amiga systems for the most part). They definitely had some rock-star case designers even back in the late 80s/early 90s. They got better at system design later on…

-Dave

[Dave Haynie]

If you install Vegas, all plug-ins, etc. as on your current PC, it’s going to be business as usual. Hopefully much faster.

If you install the 64-bit version, the project is compatible. You will need 64-bit versions of any plug-in you may be using, or you’ll lose whatever it is they do.

Both versions can be installed.

-Dave

[Dave Haynie]

[Jane Kong] “This H.265 CODEC has already become the mainstream , But still hope the codec will natively support by soem NLEs.”

“You keep using that word. I do not think it means what you think it means” -Inigo Montoya

H.265/HEVC may be the heir apparent for the next mainsteam CODEC, but it’s hardly the mainsteam just yet. Google launched the main competitor, VP9, but hasn’t had the industry support and connections of the MPEG people, so it’s pretty clearly going to become mainstream. But it’s not there yet.

It’s not yet well supported in most NLEs. Only a very few video cameras of any kind are using it yet (Samsung NX1). Netflix is using it for 4K streaming only (10-16Mb/s for a 4K stream), but only to Netflix-certified 4K televisions. YouTube is not (they’re using the rival VP9, which Google developed). Sony’s first 4K player did not support it, their new one does. Red’s doesn’t. DirecTV is supporting it, but only for 4K, and only on Samsung 4K TVs right now. The latest iOS and Android support it, but only on sufficiently capable hardware (Android 4.4 supports VP9).

At best, expect HEVC to be mainstream once 4K is mainstream.

-Dave

[Dave Haynie]

[mark thompson] “just adding my 2 cents. I’m not saying anything is wrong but I would shift the emphasis slightly.
CUDA is an architecture. You program to it in C or CUDA C which is an extension to C/libraries for C. That enables you to program the CUDA based chips from NVIDEA. “

CUDA is actually more like a language than anything else. Yes, you program it via a C API. That creates a functional graph, which will be compiled when loaded on any given nVidia GPU. You are creating a high level program when you write CUDA code. Yes, it is designed for nVidia GPUs… but their architecture has changed quite a bit over the years, while CUDA programs still run.

[mark thompson] “OpenCL is a standard that implementers can use provide a common programming interface across multiple gpu architectures.
In theory OpenCL should be a little less efficient than programming the chip’s native interface. There is no fundamental reason that you couldn’t provide an OpenCL interface on top of CUDA.”

OpenCL works much the same way, only it’s completely architecture independent. Again, there’s a run-time compiler in your OpenCL subsystem and graphics driver that targets any OpenCL program to the device of choice. OpenCL runs on GPUs, naturally. You can also run it on a CPU (AMD offers an OpenCL driver for AMD CPUs). And non-GPU compute engines — Intel’s Phi board runs OpenCL. Xilinx and Altera have OpenCL compilers for their FPGAs.

Both CUDA and OpenCL may be somewhat less efficient than programming “to the metal” on a GPU (keep in mind, CUDA isn’t directly programming a GPU either). Much like programming in C vs. assembler. On the other hand, modern CPUs are actually often more efficient when programmed in C or other HLLs, simply because a good compiler knows everything about the CPU architecture in use. It’s difficult to keep all of the various complexities of a modern CPU in one’s mind (virtual register allocation, performance of every instruction, prefetch efficiences for different instruction ordering, the various parallel integer, floating point, and special units in each CPU core that run concurrently, pipeline efficiencies and stalls, etc). CPUs are simpler processors, but more complex as a system. You can get very “low level” with CUDA, as there’s basically an assembly language mechanism available — and in fact, some developers have found that necessary to get the best performance out of Kepler processors.

As far as CUDA vs. OpenCL, your mileage may vary. CUDA may in general be a little more tightly coupled to the hardware, but it’s also largely synchronous. OpenCL is largely asynchronous, which allows it to run faster on some algorithms. And while CUDA may outperform OpenCL on some things on nVidia, it’s really difficult to get an honest comparison. AMD has long had better OpenCL implementations, and nVidia intentionally cripples some OpenCL and CUDA operations on standard and Workstation cards, since their “Compute Engine” cards also use basically the same processors. So you can find some benchmarks in which the AMD is totally clobbering the nVidia — this has been the case with Bitcoin mining, for example (well, at least back when it made economic sense to mine Bitcoins on GPUs — and curiously, it was really hard to get a top AMD GPU card for a short time as a result).

-Dave