|
Grundulum posted:Is "upgrade/change the memory" really a patentable development? No, since you can already get flash memory in a DRAM form-factor if you really need it. Never judge a patent by the title, the only part of a patent that holds legal weight is the claims -- and on the first run the claims are usually incredibly broad and some are expected to be rejected to get a feel for what the examiner considers novel. Patent lawyers consider getting an application accepted on the first try a failure because it means they could have gone even more broad and likely gotten more IP locked away.
|
# ? Aug 2, 2015 07:26 |
|
|
# ? May 8, 2024 07:21 |
|
BobHoward posted:You say there isn't a noticeable gain. How can you tell?...on what basis do you claim durability has gone down? How do you know costs haven't changed? Why do you think I'm merely theorycrafting? I have worked at a fabless semi company... LGA pins are notoriously delicate and difficult to fix even when its possible to do so, the durability difference is obvious. That costs haven't changed much is an assumption on my part. Usually if Intel has some sort of cost advantage they like to crow about it even if they won't go into specifics. I haven't seen anything like that from them and have certainly not seen anything to suggest packaging prices have changed much with LGA vs PGA. If you would've had a definitive link you probably would've posted it by now + you were giving possible issues instead of giving concrete examples with the actual existing sockets themselves. Its cool (not being ironic, it really is) you've got experience in the business but does your experience really apply to current AMD vs Intel sockets?
|
# ? Aug 2, 2015 07:36 |
|
PC LOAD LETTER posted:Where is the difference in capabilities or performance? FM2+ and LGA1150 both support high clocked dual channel DDR3 RAM, both can allow overclocks of that DDR3 RAM fairly well, both support PCIe 3, and support high clocked + high bandwidth buses. Ultimately those are all the things that actually matter and if the advantages of LGA are too small to make an impact on them then do those advantages really matter? Please go back and reread what I said about how changes in the parasitic inductance of pins can cause gains or losses in realizable clock frequency at a given TDP rating. What I have been trying to tell you all along is that real gains can be invisible if you don't know what you're looking for, or don't have the tools to measure them. You seem totally hung up on the fact that AMD got those features to work; that's not the entire picture. quote:LGA pins are notoriously delicate and difficult to fix even when its possible to do so, the durability difference is obvious. The average number of insertion cycles these sockets see is about 1, because enthusiasts are a small fraction of the market. The "durability" metrics important to Intel (and AMD) are the factory assembly scrap rates and field failure rates (within the system warranty period) observed by large PC OEMs. In that context I doubt LGA is substantially different from PGA, as once the chip is installed the delicate contacts are quite well protected.
|
# ? Aug 2, 2015 10:44 |
|
BobHoward posted:Please go back and reread what I said about how changes in the parasitic inductance of pins can cause gains or losses in realizable clock frequency at a given TDP rating. What I'm not getting is: Where is the difference between a PGA and a LGA socketed CPU wrt to conductor length and resulting inductance? I mean the pins are there in both cases, or does the decreased inductance result from something else?
|
# ? Aug 2, 2015 12:57 |
|
Grim Up North posted:What I'm not getting is: Where is the difference between a PGA and a LGA socketed CPU wrt to conductor length and resulting inductance? I mean the pins are there in both cases, or does the decreased inductance result from something else? The PGA pins have to be longer, and the geometry of those pins and the internal spring clamps gives a higher inductance than the LGA system. The pin has to stick below the side loaded spring contractor by some amount, but the LGA contact pins are also the springs, so the total conductor length is lower and the resulting pin geometry isn't as odd. Longer pins means higher inductance. When you have a 50A or 100A current surge in 30 microseconds as the chip wakes up and goes to full load, the little bit of extra inductance can cause voltage droop and overshoot. That droop and overshoot causes instability in the transistors, so you either drive the chip harder at higher voltage to correct for the undershoot, or you slow the chip down so the little bit of droop doesn't cause stability issues. You can see the same thing happen on really crappy motherboards with single phase CPU VRMs. In that case the power delivery system can't handle the changing loads, and you see the same vdroop issues and resulting weird poo poo happening. It's a pretty tiny difference, but a few nanohenries of inductance when you have a dI/dt of 5MA/s can mean a .075 or greater voltage droop for potentially 100+ microseconds, more than long enough for a 10+ clock cycles to potentially have flaky switching behavior. Now that chips are in the .9-1.1V factory voltage range, 0.075V becomes a non-trivial ripple in supply voltage. It's also why a lot of the nicer motherboards have droop compensation built in, they bump the voltage up during large load swings and bring it back down after 150ish microseconds, to cancel out the droop for the most part. It's still there, but doesn't really affect the chip thermally. Is LGA better than PGA from a pure electric performance standpoint? Yes-ish. Is it better from a cost standpoint? I don't know. Can you engineer around it with motherboard features and creative wiring? You sure as gently caress can. To the average end user, it's basically a wash, most boards are smart enough to auto-compensate for the few things that are slightly different, and aside from how you clamp it in, there isn't anything different.
|
# ? Aug 2, 2015 13:17 |
|
While I'm sure chip manufacturers don't care about this, it is significantly easier to recycle and reuse LGA CPUs vs PGA. Now you might say that just makes it harder to recycle/reuse the motherboard, but that was already crazy difficult. If a CPU is bad, it likely won't post, so if that was the point of failure, it's easy to test. Motherboards can have dozens of points of failure and are a headache no matter what socket is used, so most recyclers don't even try to reuse them. They just shred them and reclaim precious minerals, so it doesn't matter how much harder you make it to reuse motherboards.
|
# ? Aug 2, 2015 15:27 |
|
BobHoward posted:What I have been trying to tell you all along is that real gains can be invisible if you don't know what you're looking for, or don't have the tools to measure them. You seem totally hung up on the fact that AMD got those features to work; that's not the entire picture. The proof should be in the pudding so to speak and that there isn't any is very telling. I don't know a simpler way of putting things really and we're kind've going around and around on this so if its OK I'll just drop it. :/ BobHoward posted:The average number of insertion cycles these sockets see is about 1, because enthusiasts are a small fraction of the market.
|
# ? Aug 2, 2015 16:45 |
Well the good news is that CPU progress has slowed to such a glacial pace that you can just keep the same CPU/mobo for 5-10 years without any compelling reason to upgrade so plugging a new CPU into a socket isn't really something that happens anymore
|
|
# ? Aug 2, 2015 19:36 |
|
Pryor on Fire posted:Well the good news is that CPU progress has slowed to such a glacial pace that you can just keep the same CPU/mobo for 5-10 years without any compelling reason to upgrade so plugging a new CPU into a socket isn't really something that happens anymore My current PC is a P35 based mother board, bought 8 years ago. I am going to build a Skylake system this next month, that could last me 10 years or more. I wonder if the system I build after that, might be the last PC I ever build...
|
# ? Aug 2, 2015 20:57 |
|
Pryor on Fire posted:Well the good news is that CPU progress has slowed to such a glacial pace that you can just keep the same CPU/mobo for 5-10 years without any compelling reason to upgrade so plugging a new CPU into a socket isn't really something that happens anymore I think the next big improvements for computers will come outside the CPU. We are already seeing SSD's as a meaningful upgrade that is more cost efficient than a new CPU, and things like xpoint (or similar) that move faster storage closer to the CPU, as well as HBM (or similar) that move faster memory closer to the CPU will be the next big 'must haves' for computers. GPU growth and integration will continue a pace for a while, and the VR products coming in the next few years might give them a boost, but we are already heavily into the 'branding, not innovation' business model, so don't expect anything amazing.
|
# ? Aug 2, 2015 23:11 |
|
I think laptops that can be used for 10 hours strait and still be portable as pretty amazing.
|
# ? Aug 2, 2015 23:25 |
|
its almost like CPUs have made amazing advances but my gigahurtz
|
# ? Aug 2, 2015 23:29 |
|
Boiled Water posted:I think laptops that can be used for 10 hours strait and still be portable as pretty amazing.
|
# ? Aug 2, 2015 23:53 |
|
go3 posted:its almost like CPUs have made amazing advances but my gigahurtz ^^^^^^^
|
# ? Aug 3, 2015 00:12 |
|
Except that they haven't? Or ? http://www.techspot.com/article/1039-ten-years-intel-cpu-compared/page3.html Unless the application uses AVX2 or something, the difference from C2Q era to the 4790K is like 2-3 times. Which is nice but kind of lame really for 8 years of development.
|
# ? Aug 3, 2015 17:31 |
|
then compare performance/watt, which has gotten a lot better?
|
# ? Aug 3, 2015 17:34 |
|
kujeger posted:then compare performance/watt, which has gotten a lot better? I think the point is in reasonable raw day to day performance, a metric that most enthusiasts care far more about than saving pennies on an electrical bill.
|
# ? Aug 3, 2015 17:37 |
|
PerrineClostermann posted:I think the point is in reasonable raw day to day performance, a metric that most enthusiasts care far more about than saving pennies on an electrical bill. Performance / watt has enabled modern laptops to become what they are. It's pretty incredible, especially as gaming workloads have consistently remained 100% GPU limited.
|
# ? Aug 3, 2015 17:44 |
|
Twerk from Home posted:Performance / watt has enabled modern laptops to become what they are. It's pretty incredible, especially as gaming workloads have consistently remained 100% GPU limited. Not arguing that, my t100 is great. But most enthusiasts do their heavy lifting on their towers, not on a more expensive, less able laptop. So again, that's not the metric enthusiasts particularly care for.
|
# ? Aug 3, 2015 17:48 |
|
I think he's saying he'd rather have a worse SuperPi score with a 7ghz CPU vs a better SuperPi score with a 3ghz CPU. Also looking at the SuperPi records makes me long for better cooling. 7Ghz or go home. http://hwbot.org/benchmark/superpi_-_1m/
|
# ? Aug 3, 2015 17:51 |
|
.
sincx fucked around with this message at 05:55 on Mar 23, 2021 |
# ? Aug 3, 2015 17:59 |
|
PerrineClostermann posted:Not arguing that, my t100 is great. But most enthusiasts do their heavy lifting on their towers, not on a more expensive, less able laptop. Physics is a bitch. Until we can figure out an entirely new mode of computing, we'll have to focus on parallelizing our work instead of just making one really fast core. Cheap space travel would be nice too, but it turns out, it is actually REALLY HARD.
|
# ? Aug 3, 2015 18:19 |
|
kujeger posted:then compare performance/watt, which has gotten a lot better? For laptop parts, maybe. My Q6600 has a TPD of 105W vs 88W for 4790K. Again, that's nice, but kinda lame. http://ark.intel.com/products/29765/Intel-Core2-Quad-Processor-Q6600-8M-Cache-2_40-GHz-1066-MHz-FSB http://ark.intel.com/products/80807/Intel-Core-i7-4790K-Processor-8M-Cache-up-to-4_40-GHz
|
# ? Aug 3, 2015 18:38 |
|
PerrineClostermann posted:I think the point is in reasonable raw day to day performance, a metric that most enthusiasts care far more about than saving pennies on an electrical bill. enthusiasts are pretty irrelevant so i'm not particularly sure why Intel should be catering to them
|
# ? Aug 3, 2015 18:46 |
|
true performance nerds buy xeon cluster farms
|
# ? Aug 3, 2015 18:48 |
|
mobby_6kl posted:For laptop parts, maybe. My Q6600 has a TPD of 105W vs 88W for 4790K. Again, that's nice, but kinda lame. Also, isn't it the case that the TDPs for Haswell were UP on Ivy Bridge? (Same clocks). I guess it was something to do with integrating the voltage regulator.. HalloKitty fucked around with this message at 19:35 on Aug 3, 2015 |
# ? Aug 3, 2015 19:30 |
|
Skandranon posted:Physics is a bitch. Until we can figure out an entirely new mode of computing, we'll have to focus on parallelizing our work instead of just making one really fast core. Cheap space travel would be nice too, but it turns out, it is actually REALLY HARD. Not arguing against that truth either. go3 posted:enthusiasts are pretty irrelevant so i'm not particularly sure why Intel should be catering to them Or that.
|
# ? Aug 3, 2015 19:36 |
|
sincx posted:Desktop performance--when not power limited and using reasonable overclocks--has barely improved (if at all) over the last few years because of max clock speeds getting worse and worse, thus cancelling out any IPC gains. It looks like Intel just made their variability a lot better (i.e., smaller) so a "great" overclock is much closer to the mean.
|
# ? Aug 3, 2015 19:39 |
|
mobby_6kl posted:For laptop parts, maybe. My Q6600 has a TPD of 105W vs 88W for 4790K. Again, that's nice, but kinda lame. That's just watts, not performance per watt. Picking a random benchmark from Anandtech's comparisons: Cinebench single-threaded: Q6600: 2778 CBMarks; 26.46 CBMarks per watt i7-4790K : 8785 CBMarks; 99.8 CBMarks per watt That's nearly a fourfold improvement. I admit, the gains in a given year don't seem especially compelling anymore to me either, but it's not like there hasn't been noticeable progress.
|
# ? Aug 3, 2015 19:43 |
|
Skylake packaging is uh... huh http://hexus.net/tech/news/cpu/85187-intel-skylake-core-i7-6700k-i5-6600k-packaging-leaks-online/ http://www.overclock3d.net/articles/cpu_mainboard/intel_skylake_box_pictured_k-cpus_will_not_come_with_cpu_coolers/1 Also confirmed that they don't come with any CPU coolers at all
|
# ? Aug 3, 2015 23:32 |
|
Cardboard Box A posted:Skylake packaging is uh... huh
|
# ? Aug 3, 2015 23:44 |
|
Marinmo posted:Oh god the thread will hate me but I actually don't think they look bad at all That's a perfectly fine opinion to have, since it's a CPU box. It's something you see for a few moments, take the CPU out of, and then throw back in the case box with all the other boxes of poo poo you just unpacked to build your new system. You know you have nothing to give a poo poo about when you're thinking about the graphic design of a CPU box.
|
# ? Aug 4, 2015 00:03 |
|
Cardboard Box A posted:Skylake packaging is uh... huh I'm ok with this
|
# ? Aug 4, 2015 00:04 |
|
Intel seems to care a little bit too much about the design of something that people hardly ever see because: 1) They are getting that poo poo by the tray 2) Already bought it off Newegg before seeing anything so at that point who gives a gently caress what the box looks like. And if the goal is to recreate an Apple unboxing experience then
|
# ? Aug 4, 2015 00:13 |
|
I refuse to buy a Skylake over this #NotMYskylake
|
# ? Aug 4, 2015 00:27 |
|
Holy gently caress, I better wait for Zen.
|
# ? Aug 4, 2015 00:36 |
|
wipeout posted:Holy gently caress, I better wait for Zen.
|
# ? Aug 4, 2015 00:52 |
|
japtor posted:Hopefully they get inspired by the old GPU boxes, like have some monk meditating with a bunch of video gamey poo poo flying out of his head. Sadly, I can't find a picture of the 3D Nuclear Pope XL box.
|
# ? Aug 4, 2015 01:29 |
|
japtor posted:Hopefully they get inspired by the old GPU boxes, like have some monk meditating with a bunch of video gamey poo poo flying out of his head. Intel can have a sexy blue-haired anime babe (holding a gun) on their CPU boxes and AMD can have a sexy red-haired robot babe (holding a gun) on their CPU boxes.
|
# ? Aug 4, 2015 04:21 |
|
|
# ? May 8, 2024 07:21 |
|
Are we too far past a dark blue background with 3D geometric shapes of various primary and secondary colors hovering over a purple-blue grid plane?
|
# ? Aug 4, 2015 04:35 |