|
Alereon posted:Significantly, yes. Is it possible to disable the graphics card in your laptop? Check the catalyst control centre for support. The NVidia drivers running on my laptop allow specifying the graphic conditions for using the intel or amd gpus. If not then there should be a default gpu option.
|
#
?
Oct 27, 2011 10:15
|
|
|
|
| # ? Dec 13, 2024 13:18 |
|
|
Hey since we have all these faster processors cranking out, how does a 8 core affect gaming? do half the mmo's out there and pc games even have coding to use a 8 core ?
|
|
#
?
Oct 27, 2011 12:34
|
|
|
Far from universally. In the world of MMOs, World of Warcraft creaks in at using 2 threads, I believe, one for the game, one for the audio. Crysis didn't scale meaningfully past two cores, if I recall correctly. The first game I recall that abused many cores was Supreme Commander. I've seen GTA IV dole out abuse to at least 3 cores. You'll see a bit more of that as time goes by, because the Xbox 360 has 3 CPU cores.. If it was a stronger chip, we would have good 6-way scaling by now, since each 360 core is HT capable. Still, the key seems to be performance per core if you're looking at gaming. Which AMD fails at, even though they're trying to target gamers. HalloKitty fucked around with this message at 12:56 on Oct 27, 2011 |
|
#
?
Oct 27, 2011 12:53
|
|
|
streetgang posted:Hey since we have all these faster processors cranking out, how does a 8 core affect gaming? do half the mmo's out there and pc games even have coding to use a 8 core ? I think all the benchmarks have shown "8" core bulldozer chips losing handily to 4 core Sandy Bridge and even pre-Sandy Bridge chips in games.
|
|
#
?
Oct 27, 2011 14:25
|
|
|
streetgang posted:Hey since we have all these faster processors cranking out, how does a 8 core affect gaming? do half the mmo's out there and pc games even have coding to use a 8 core ? HalloKitty pretty much covered it. Biggest applications for multiple cores are in productivity applications. Developers will see faster compile times (we use one of our old 32-core Opteron chassis to compile/simulate VHDL/Verilog), certain server applications like databases prefer cores to raw clock speed, etc. VMs also of course benefit from many cores, and it's nice to get a large # of cores while consuming less rack units. Multiple cores are definitely awesome, but all you need for gaming at the moment is a quad-core.
|
|
#
?
Oct 27, 2011 14:44
|
|
|
God drat it why do I feel like I'll end up going from Athlon64 -> Piledriver based on the name alone.  I've actually been meaning to ask: where do these companies get the names for their processors?
|
|
#
?
Oct 28, 2011 04:01
|
|
|
angrytech posted:God drat it why do I feel like I'll end up going from Athlon64 -> Piledriver based on the name alone. Intel uses a map. quote:Intel has historically named integrated circuit (IC) development projects after geographical names of towns, rivers or mountains near the location of the Intel facility responsible for the IC. Many of these are in the American West, particularly in the state of Oregon (where most of Intel's CPU projects are designed; see well-known project codenames). As Intel's development activities have expanded, this nomenclature has expanded to Israel and India. Some older codenames refer to celestial bodies. There is a pattern with recent desktop processors. Since Core 2 all quad-core desktop processors tend to end in "field" (e.g. Kentsfield, Bloomfield, Lynnfield) and most desktop dual-cores end in "dale" (e.g. Wolfdale, Allendale, Clarksdale), with the exception of Arrandale, a mobile processor codename for the mobile i3/i5/i7s. Platforms consisting of a CPU plus a Southbridge end in "trail" (e.g. Bone Trail, Skull Trail, Pine Trail). Server processors for two sockets now end in "town" (e.g. Harpertown, Gainestown, Gulftown), while server processors for four or more sockets end in "ton" (Tigerton, Dunnington, Beckton).
|
|
#
?
Oct 28, 2011 04:07
|
|
|
angrytech posted:God drat it why do I feel like I'll end up going from Athlon64 -> Piledriver based on the name alone. http://www.10stripe.com/articles/where-do-they-get-their-codenames.php http://en.wikipedia.org/wiki/List_of_Intel_codenames I'd lmgtfy it, but 
|
|
#
?
Oct 28, 2011 04:11
|
|
|
The real tragedy here is that Bulldozer is such an awesome code name and it has been utterly wasted.
|
|
#
?
Oct 28, 2011 04:13
|
|
|
Galler posted:The real tragedy here is that Bulldozer is such an awesome code name and it has been utterly wasted. Bulldozer was simply foreshadowing what would be needed to get rid of the drat things.
|
|
#
?
Oct 28, 2011 04:42
|
|
|
Galler posted:The real tragedy here is that Bulldozer is such an awesome code name and it has been utterly wasted.
|
|
#
?
Oct 28, 2011 04:43
|
|
|
Anandtech has a small article on AMD's Q3 results, they're basically being carried on the backs of their Fusion processors (especially Llano), which continue to sell as fast as they can make them. Average Selling Price (ASP) remains low, because nobody buys high-end AMD processors. They saw growth in the mobile and server sectors. The graphics division is only slightly profitable, though volumes are up for both desktop and mobile and ASP is also up (since nobody buys LOW-END graphics cards now). Next quarter we get Q4 results and see what happened when they launched Bulldozer, and whether it's got any traction in the server space. Oddly (given the disappointing desktop performance), there's been some announcements of supercomputers going with Bulldozer processors, and I sincerely doubt they wouldn't THOROUGHLY test them to see what kind of performance and power usage they can expect in their applications before signing a contract for thousands of processors. It may just be that 16 cores per socket, AMD's memory advantages (huge arrays of LRDIMMs mean you can get a shitton of RAM cheaply and with very low power usage compared to Intel), and configurable TDPs are too compelling.
|
|
#
?
Oct 28, 2011 07:57
|
|
|
TechReport has done a quick test on Bulldozer thread scheduling. It looks like they had the same idea I had, where you set up different patterns of core affinities and see what happens.
|
|
#
?
Oct 28, 2011 13:10
|
|
|
Maxwell Adams posted:TechReport has done a quick test on Bulldozer thread scheduling. It looks like they had the same idea I had, where you set up different patterns of core affinities and see what happens. Okay, so when will Tom's or Anand re-run their benchmarks with these options? Are there any cases where disabling HT on the latest Intel processors increases performance?
|
|
#
?
Oct 28, 2011 13:37
|
|
|
Bob Morales posted:Are there any cases where disabling HT on the latest Intel processors increases performance? If you're still running XP SP2 or earlier, the OS' scheduler has problems handling hyperthreading correctly, if I remember right. An edge case to be sure.
|
|
#
?
Oct 28, 2011 14:07
|
|
|
trandorian posted:If you're still running XP SP2 or earlier, the OS' scheduler has problems handling hyperthreading correctly, if I remember right. An edge case to be sure. I thought there was something about the first Pentium 4's with HT's basically performed like poo poo on single-threaded stuff because the cache was effectively split in half.
|
|
#
?
Oct 28, 2011 14:09
|
|
|
Maxwell Adams posted:TechReport has done a quick test on Bulldozer thread scheduling. It looks like they had the same idea I had, where you set up different patterns of core affinities and see what happens. That article is a bit deceptive, because there's one thing it doesn't make clear... all of those scores are significantly lower than if they were just allowed to run with 8 threads in the first place. The other thing that's not clear is to what extent those benchmarks are floating point. With FP operations, the modules really are a lot closer to simply a single core with Hyperthreading.
|
|
#
?
Oct 28, 2011 14:37
|
|
|
Alereon posted:The graphics division is only slightly profitable, though volumes are up for both desktop and mobile and ASP is also up (since nobody buys LOW-END graphics cards now). It's a little sad the graphics division isn't too profitable, they've been absolutely killing it for years now.
|
|
#
?
Oct 28, 2011 16:13
|
|
|
Zhentar posted:That article is a bit deceptive, because there's one thing it doesn't make clear... all of those scores are significantly lower than if they were just allowed to run with 8 threads in the first place. I think it's not trying to compare 2/4 thread vs. 8 thread, it's figuring out how to best schedule when there's not full core/module saturation. Like in games. If windows is trying to maximize idle cores/modules while in lightly threaded situations it could lead to lower performance. This might be what the windows 8 10% performance increase comes from. Longinus00 fucked around with this message at 16:32 on Oct 28, 2011 |
|
#
?
Oct 28, 2011 16:30
|
|
|
quote:distributing the threads one per module, and thus avoiding sharing, produces roughly 10-20% higher performance than packing the threads together on two modules. bulldozer was already rocking the house with multi-threaded integer arithmetic, so if applications and/or operating systems can just figure out how to keep FP arithmetic from sharing cores, we'll be in business. I need another 6% jump in AMD stock to be profitable again.
|
|
#
?
Oct 28, 2011 16:30
|
|
|
Mathhole posted:bulldozer was already rocking the house with multi-threaded integer arithmetic, so if applications and/or operating systems can just figure out how to keep FP arithmetic from sharing cores, we'll be in business. I got out of it the other day, then it decided to jump up 20% 
|
|
#
?
Oct 28, 2011 16:33
|
|
|
angrytech posted:God drat it why do I feel like I'll end up going from Athlon64 -> Piledriver based on the name alone. For a while in the mid 90's, AMD codenamed their processors after Land Before Time characters (Chompers, Little Foot, Sharptooth).
|
|
#
?
Oct 28, 2011 17:10
|
|
|
Ugh.  Why can't it be like this all the time? 
|
|
#
?
Oct 28, 2011 17:48
|
|
|
Longinus00 posted:I think it's not trying to compare 2/4 thread vs. 8 thread, it's figuring out how to best schedule when there's not full core/module saturation. Like in games. If windows is trying to maximize idle cores/modules while in lightly threaded situations it could lead to lower performance. This might be what the windows 8 10% performance increase comes from. Yeah, I realize that's what they're intending to compare, but the article doesn't do a good job of conveying that; I was pointing it out because it would be easy for someone to walk away from that article with the wrong conclusion. My other complain about not being clear about the floating-pointness is because that directly impacts how applicable it is to games and other desktop scenarios. If it's faster because it reduces contention on FP components, then it's meaningless for most desktop workloads. If it's faster because it reduces cache contention, or some other reason, then it's more likely to help other workloads.
|
|
#
?
Oct 28, 2011 22:42
|
|
|
Zhentar posted:Yeah, I realize that's what they're intending to compare, but the article doesn't do a good job of conveying that; I was pointing it out because it would be easy for someone to walk away from that article with the wrong conclusion. It might help out even in non FP situations because BD shares decoders across a module. You might get better performance simply by being able to throw all of a modules decoders at one thread instead of two. What this is going to power consumption is a different matter.
|
|
#
?
Oct 28, 2011 23:52
|
|
|
HardOCP has an article up comparing overclocked gaming performance between Bulldozer and an i5 2500K. Unsurprisingly, Bulldozer loses, badly. This should shut up those few fanboys who rave about how great it overclocks.
|
|
#
?
Nov 3, 2011 21:16
|
|
|
The performance in ARMA 2 pretty well sums up bulldozer's capabilities.
|
|
#
?
Nov 3, 2011 21:33
|
|
|
AMD is going to layoff 1400 workers http://www.marketwatch.com/story/amd-to-ax-1400-workers-in-cost-cutting-move-2011-11-03?link=MW_latest_news
|
|
#
?
Nov 3, 2011 21:42
|
|
|
Alereon posted:HardOCP has an article up comparing overclocked gaming performance between Bulldozer and an i5 2500K. Unsurprisingly, Bulldozer loses, badly. This should shut up those few fanboys who rave about how great it overclocks. Worse, Bulldozer looks almost unplayable in most of the graphs. It spikes up and down like mad.. that would be a horrible experience. In one of the tests, the 2500k has a minimum of about 15fps - not great, but not unplayable, where Bulldozer hits <5fps, which would be horrendous.
|
|
#
?
Nov 3, 2011 22:13
|
|
|
I was surprised to see that the mean performance delta seemed pretty commensurate between the Arma tests in particular - but then you see the crazy, crazy, crazy graph and realize that a trend line doesn't mean much when your data points are jumping around like crazy. AMD firing 1400 people? Where are those jobs coming from? Ugh I hate the stock market, take a bunch of people and gently caress their lives up and stock goes up. Looks like they will be pursuing the K10.5 thing after all, though? Maybe good news? Edit: Oh, ARM. Great. Agreed fucked around with this message at 22:51 on Nov 3, 2011 |
|
#
?
Nov 3, 2011 22:49
|
|
|
Agreed posted:Edit: Oh, ARM. Great. The ARM part is just some dipshit analyst speculating that their plans to focus on "lower power" means they're going to use ARM. As opposed to, say, investing in their product line that's actually doing well.
|
|
#
?
Nov 4, 2011 01:07
|
|
|
Zhentar posted:The ARM part is just some dipshit analyst speculating that their plans to focus on "lower power" means they're going to use ARM. As opposed to, say, investing in their product line that's actually doing well.
|
|
#
?
Nov 4, 2011 01:32
|
|
|
Alereon posted:HardOCP has an article up comparing overclocked gaming performance between Bulldozer and an i5 2500K. Unsurprisingly, Bulldozer loses, badly. This should shut up those few fanboys who rave about how great it overclocks. It makes no sense. If you need to spend $100+ on WC to overclock BD to beat a stock SB why not just GODDAMN buy the better 2500/2600K in the first place that OCs sky high on a $30 Hyper 212+?
|
|
#
?
Nov 4, 2011 01:48
|
|
|
freeforumuser posted:It makes no sense. If you need to spend $100+ on WC to overclock BD to beat a stock SB why not just GODDAMN buy the better 2500/2600K in the first place that OCs sky high on a $30 Hyper 212+? Because if no one buys those Bulldozer CPUs just imagine how much Intel is going to start charging for theirs! You can thank me when you pick up a cheap new Ivy Bridge processor next spring. 
|
|
#
?
Nov 4, 2011 02:01
|
|
|
BEAR GRYLLZ posted:Because if no one buys those Bulldozer CPUs just imagine how much Intel is going to start charging for theirs! You can thank me when you pick up a cheap new Ivy Bridge processor next spring. I'd prefer if you bought a Sandy Bridge and paypal'd me the money you save every month on your electricity bills.
|
|
#
?
Nov 4, 2011 16:16
|
|
|
HalloKitty posted:Worse, Bulldozer looks almost unplayable in most of the graphs. Bulldozer pales in comparison to the 2500k, but realize that a lot of those benchmarks were done on a triple-head setup, and most of us won't be gaming on those kind of setups. They even do a "playable" summary, which shows that for most people, BD is adequate for gaming. You'd still be a frothing fanboy to purchase AMD right now, though, all other things being equal.
|
|
#
?
Nov 4, 2011 16:58
|
|
|
Civil posted:You'd still be a frothing fanboy to purchase AMD right now, though, all other things being equal. I wouldn't go that far, you'd be a frothing fanboy to purchase Bulldozer though. The 955/960 still have merit as being affordable adequate chips.
|
|
#
?
Nov 5, 2011 13:22
|
|
|
Alereon posted:HardOCP has an article up comparing overclocked gaming performance between Bulldozer and an i5 2500K. Unsurprisingly, Bulldozer loses, badly. This should shut up those few fanboys who rave about how great it overclocks.
|
|
#
?
Nov 5, 2011 16:05
|
|
|
Combat Pretzel posted:Holy crap. Are they going to blame that on scheduler problems? I wonder how that ARMA benchmark would look like with a 2600K in play, for 4C8T SMT. Well, it probably IS a scheduler problem isn't it? Fixing it of course would probably only smooth out the spikes in both directions though, not increase overall performance.
|
|
#
?
Nov 5, 2011 16:30
|
|
|
|
| # ? Dec 13, 2024 13:18 |
|
|
Install Gentoo posted:Well, it probably IS a scheduler problem isn't it? Fixing it of course would probably only smooth out the spikes in both directions though, not increase overall performance.
|
|
#
?
Nov 5, 2011 22:44
|
|