PC LOAD LETTER posted:

edit:/\/\/\/\Yea looks like Intel's patent licenses are safe no matter what if AMD goes under I'm reading that right.

Nah, they'd just buy up whatever patents they wanted. Not many can out bid Intel or are even interested in buying up those patents. Even then the details of any license agreement, and the method of bankruptcy, between AMD and Intel may mean that Intel won't necessarily lose access to them even if AMD goes under and someone else gets them. For all we know AMD deeded Intel the patents in the event of bankruptcy as part of a patent sharing contract.

There is a reason why any bankruptcy would probably be a lengthy affair folks. The lawyers could be hashing stuff out for years.


edit: \/\/\/\/\/\/Well sure only a select few investors/insiders/good 'ol boys benefit from any sort of monopoly but that group tends to be pretty powerful and gets what they want. Particularly in today's political/regulatory climate. I'm not advocating their position, I'm just pointing out that is the way it is right now. ARM might indeed be able to push Intel/x86 out of the market in a decade or so but that is a whole other subject altogether and isn't something that is a sealed deal yet either. Speaking about these things as if they're foregone conclusions is premature to say the least.

Alereon posted:

If there's a change in ownership of either company, the cross-licensing agreement is terminated for both Intel and AMD. Intel can't decline a new agreement with AMD's new owner because they need the IP to make 64-bit CPUs, and I think their unique relationship with MS would keep them from pushing too hard on licensing terms.

Comedy option: Microsoft buys AMD, Apple buys Intel. Everybody else in Europe, Africa, and most of Asia buys ARM CPUs in Android devices, and China keeps going MIPS on Linux.

Cardboard Box A posted:

So uh basically going from 120Hz to 144Hz on a high refresh rate monitor is enough to kick NVIDIA's GPUs from their low power state into their full power state when just on the Windows Desktop, no games....

monster on a stick posted:

I'm in a similar position - I'd rather not upgrade anything else right now unless I need to, so what's a good card to look for that will play TW3 but take up about the same amount of power? Would an R9 380 work?

THE DOG HOUSE posted:

GrizzlyCow posted:

Is there a reason all the good IPS(-like) *Sync displays are having so much problems? The TN panels aren't suffering these issues, right?

slidebite posted:

Sorry if this is a stupid question, but is Geforce Experience a thing to have? I always just install the basic drivers and really nothing else. Am I missing anything?

causticBeet posted:

Ended up getting an insignia 520w psu at bestbuy because I overlooked that (dope).

Police Automaton posted:

I use an HD6450 as second graphics card and it does 1080p60 youtube fine in firefox, and that's in Linux. It doesn't even use video hardware acceleration by the graphics card, I'm not even sure this is possible in my particular configuration.

ufarn posted:

(Speaking of, why do the videos annihilate my CPU? Shouldn't my R9 270X GPU bear some of that brunt?)

Fauxtool posted:

what are my options when it comes to making 2 monitors look similar. Can a calibrator still help even if 1 is ips and the other TN?

sauer kraut posted:

I don't think youtube VP9 videos are hardware decoded unless you play it on a newer Intel integrated GPU.
1080p/60fps can be pretty taxing.

HalloKitty posted:

XFX is now good and Sapphire is poo poo? It seems like the answer to that question hangs on whoever you ask that day.

I'd put Sapphire above XFX, for what it's worth.

Naffer posted:

You're making me feel old. My first 3d card was a 3dfx voodoo banshee. It could get 60fps at quake 2 at 640x480! (http://www.tomshardware.com/reviews/3d-chips,83-7.html)
The MX cards were a strange beast. The Ti4200 was the best deal of that generation (and incidentally, my third video card).

Subjunctive posted:

No, it only terminates the rights of the purchased party.

Fatal posted:

K well, not at that price point but the idea still stands. I would really be interested to know how something like this would actually sell. How many people upgrade from a 970 to a 980Ti? Or even better question, how many people that upgrade stay with the same vendor?

I guess this is kinda like the evga step up program without the inconvenience of actually shipping hardware. It seems like the real risk would be to the manufacturer if somebody figured out how to unlock the card without paying for it.

SwissArmyDruid posted:

Speaking of Nvidia, what's this I'm hearing about drivers bricking cards? I thought Nvidia was supposed to be the company with the *good* driver team?

Truga posted:

I tried doing this and it wouldn't work, because ASUS in all their knowledge decided not to wire the onboard gpu to anything at all, probably thinking "this is an xfire/sli mobo, surely nobody will use the haswell gpu for anything??".

GokieKS posted:

What? No, the cards listed as using "DDR3" are actually using plain old DDR3, not GDDR3 - If you look at the VRAM chips on any of the (terrible) nVidia cards with DDR3 they keep rebadging and search the part numbers, you'll see they're actual DDR3 modules.

FaustianQ posted:

So wait, would DDR4 in fact make a decent enough performance boost over DDR3? Say a 2GB DRR4 @ 4000mhz card with a 128 bit bus, I'm likely miscalculating but that should be ~64GB/s? There should be at least some PCB area and power savings, right?

FaustianQ posted:

Well, the original thought experiment was that since you'd have a limit to financial return on heavily put P11 dies, AMD would fill out the lowest end of the 400 series lineup by lifting APU iGPUs and putting them on PCB. For potentially minimal fuss they'd use DDR4, as a different memory standard would require redesigning. As pointed out by Paul, the DDR4 memory controller being attached to the CPU could complicate the task and you'd end up at square one producing two separate but functionally identical performance GPUs.

Paul MaudDib posted:

XDMA CrossFire is absolutely bulletproof and significantly outperforms SLI in every way - frame pacing, scaling, and physical durability.

Subjunctive posted:

Is there a way to programmatically tell that the driver is in that mode? I've been entertaining writing a warning overlay/widget.

EdEddnEddy posted:

And this is pretty much exactly what caused all the lovely electronics issues of 2005 - 2009~ish. Removal of Lead in the solder made the replacement crap brittle after a bunch of heat cycles which lead to the RLOD/YLOD and all the GPU and even Auto ECU deaths over those years.

SwissArmyDruid posted:

The wrongness of this post has caused me to write and delete this post three times just trying to write an informative and enlightening post about the lead-free solder switchover, and my time working in semiconductor manufacturing during that time. It proceeds to get angrier and angrier with length, so we'll do this short:

The lead-free solder switchover was a real thing. Solders that purported to be drop-in replacements weren't quite, and needed to be cooked a little hotter to really get melty. In cases where you saw widespread solder failures, it was because they didn't get cooked hot enough.

SwissArmyDruid posted:

In short: More pixels mean less jaggies, because each individual pixel is a much smaller part of an overall line or edge, so the "stair effect" is reduced. This means you can turn AA down or off entirely and reclaim the power you'd normally use for it. This power can then be put towards texture quality or effects or whatever.

mobby_6kl posted:

Seems like it would work exactly like supersampling though, except that instead of taking an average and discording the original pixels, they're all displayed. You just can't see them individually because they're too small, assuming a high enough DPI. Which is a problem if your 4k screen is >40", then you're hosed

Larrymer posted:

Taima posted:

Grog posted:

The only comparison I know of offhand is the Digital Foundry one with a 2500K and 3770K compared to a 6500: http://www.eurogamer.net/articles/digitalfoundry-2016-is-it-finally-time-to-upgrade-your-core-i5-2500k. It seems like increased memory speed has a bigger effect on Sandy Bridge, with relatively minor gains on Ivy for certain games. The minimums increase a decent amount in some cases, but it seems like memory speed doesn't make a huge improvement on their Ivy setup.

There are still pretty big gains in minimum frame rates going from the overclocked 2500K and DDR3-2133 to the other CPUs, though. I'm still on a 2500K at 4.5 GHz, DDR3-1600, and a GTX 770. All of this is starting to make me really feel the itch to upgrade...

Watermelon Daiquiri posted:

Wouldn't 110 be, you know, worse than 108? Itd be like 1120 poo poo tier. And i think that 14nm for the next gen is mixing rumors up......

Geemer posted:

Apparently the new Geforce Experience 3.0 requires you to log into a Nvidia, Facebook or Google account. Which loving rear end in a top hat thought that was a good idea?

Guess I'm not going to experience the new Geforce Experience.