|
ItBreathes posted:I wonder if I could start selling audiophile woo to gamers? KillerPCIe 
|
#
?
Mar 23, 2019 15:31
|
|
|
|
| # ? May 17, 2024 15:51 |
|
|
ConanTheLibrarian posted:KillerPCIe Ahh good another piece of hardware for enthusiasts to remove / work around a la killer wireless I love a good pointless tech project ... Unironically
|
|
#
?
Mar 23, 2019 15:45
|
|
|
Hey kids. You wanna buy some soundcards?
|
|
#
?
Mar 23, 2019 18:25
|
|
|
Only if you also have some incredibly cheap audio cables that I can twist around all my other wires for maximum interference.
|
|
#
?
Mar 23, 2019 18:57
|
|
|
ConanTheLibrarian posted:KillerPCIe Oh, KillerPCIe definitely exists, it's just made by HIS 
|
|
#
?
Mar 23, 2019 23:21
|
|
|
I'd like to hear what these people thought this thing would do, if their intent wasn't maliciously ripping off people. --edit: Trying to find more info about the thing, I ran into a more hilarious version. 
|
|
#
?
Mar 24, 2019 00:14
|
|
|
Combat Pretzel posted:I'd like to hear what these people thought this thing would do, if their intent wasn't maliciously ripping off people. Because it just has a ton of capacitors I'm guessing it's claiming to provide "enhanced power stability" or something. I guess in theory it could make a difference under heavy load, alleviating voltage dips when there was a sudden spike in power demand, but CPUs and graphics cards should already be designed to run within the ATX power supply specs.
|
|
#
?
Mar 24, 2019 00:44
|
|
|
The board with more caps literally has some of each major type. It's like they went to the digikey cap section and picked cool looking versions of each.
|
|
#
?
Mar 24, 2019 01:10
|
|
|
https://www.reddit.com/r/Amd/comments/b4q01o/a_quick_analysis_of_the_new_zen_2_bios_options does this seem to make sense?
|
|
#
?
Mar 24, 2019 03:30
|
|
|
.
sincx fucked around with this message at 05:50 on Mar 23, 2021 |
|
#
?
Mar 24, 2019 03:42
|
|
|
sincx posted:This sounds like a really bad idea. Being substantially closer to the actual power drawing component could have non-snakeoil effects on power quality. Measurable and possibly even useful in the edge cases, but useless 95% of the time.
|
|
#
?
Mar 24, 2019 04:10
|
|
|
Methylethylaldehyde posted:Measurable and possibly even useful in the edge cases, but useless 95% of the time. Much like 8+ phase VRMs on consumer platforms 
|
|
#
?
Mar 24, 2019 07:00
|
|
|
EmpyreanFlux posted:I'm kinda confused on how your 200GE is doing better but that might be purely game selection. A tuned FX-8350 (4.5Ghz core, 300Mhz FSB, 2700Mhz NB, 2400Mhz DDR3) will beat a 200GE in all but the most single threaded of applications, even when the 200GE is overclocked. Anything needing 4 threads and there is a noticeable difference in the FX 8350s favor. I'm sure you're right and it just could be placebo, but I have it at stock speeds and with 2100 DDR3 I think. It is probably also game selection, as I don't think I play much that uses the cores the fx has. I remember my single core benchmark being faster on the 200GE, but I could have gotten something mixed up in that process.
|
|
#
?
Mar 24, 2019 11:26
|
|
|
Darth Llama posted:I'm sure you're right and it just could be placebo, but I have it at stock speeds and with 2100 DDR3 I think. It is probably also game selection, as I don't think I play much that uses the cores the fx has. I remember my single core benchmark being faster on the 200GE, but I could have gotten something mixed up in that process. Stock config for an FX processor depends, they'll do 2.4ghz stock on NB in a 970 or 760G board and that can noticeably reduce performance. The construction cores were a mess. Single core the 200GE should dominate the FX noticeably, should be around 150 in CB15 for the 200GE while the FX in the best configuration will struggle to break 115.
|
|
#
?
Mar 24, 2019 14:22
|
|
|
Mr.Radar posted:Because it just has a ton of capacitors I'm guessing it's claiming to provide "enhanced power stability" or something. I guess in theory it could make a difference under heavy load, alleviating voltage dips when there was a sudden spike in power demand, but CPUs and graphics cards should already be designed to run within the ATX power supply specs. It kinda reminds me of a quack gadget I saw in a device teardown video on youtube of a thing marketed as a means to reduce your electricity bill. IIRC the idea was you plug the box into an ac socket and the capacitors inside do black magic with PFC (which I don't think consumers actually get billed for / is more an industrial plant level of electricity usage concern?)
|
|
#
?
Mar 24, 2019 15:18
|
|
|
Bloody Antlers posted:It kinda reminds me of a quack gadget I saw in a device teardown video on youtube of a thing marketed as a means to reduce your electricity bill. IIRC the idea was you plug the box into an ac socket and the capacitors inside do black magic with PFC (which I don't think consumers actually get billed for / is more an industrial plant level of electricity usage concern?) Worse its more audiophile snake oil, an Elfidelity Power Filter Card. Filter your pci/pcie (it's double sided!) power for smooth noise free audio. They also make fan, sata, and usb versions. Still haven't found any indication it does anything since most people are doing multiple more sensible things in addition.
|
|
#
?
Mar 24, 2019 15:53
|
|
|
EmpyreanFlux posted:Stock config for an FX processor depends, they'll do 2.4ghz stock on NB in a 970 or 760G board and that can noticeably reduce performance. The construction cores were a mess. https://cpu.userbenchmark.com/Compare/AMD-FX-8350-vs-AMD-Athlon-200GE/1489vsm592714 In addition to single core advantage, which matters more for many games, the 200GE has much better floating point performance. The motherboard and RAM probably have positive performance effects as well. If you overclock, the 200GE does better still!
|
|
#
?
Mar 24, 2019 17:00
|
|
|
Methylethylaldehyde posted:To be fair, those are really tiny capacitors, and the inrush current needed to completely fill them is basically a blip on the 150w pcie bus power system. I mean, that's sorta true, but those capacitors are almost as far from another card as the other card is from capacitors on the mainboard. Any well designed card should have plenty of capacitance on-board and very close to the ICs, anyway. I vote stupid bullshit, but maybe useful if all your other parts are incredibly borderline cheap poo poo.
|
|
#
?
Mar 25, 2019 00:32
|
|
|
Unhappy Meal posted:Worse its more audiophile snake oil, an Elfidelity Power Filter Card. Filter your pci/pcie (it's double sided!) power for smooth noise free audio. They also make fan, sata, and usb versions. Still haven't found any indication it does anything since most people are doing multiple more sensible things in addition. Audiophile thought process: Gotta spend big $$$ to solve non-problems that doesn't exist outside of extreme worst case scenarios.
|
|
#
?
Mar 25, 2019 02:19
|
|
|
LRADIKAL posted:I'm sure you know, but further down the line pcie 4 may cheapen boards due to less lanes needed for save bandwidth. In the other direction, you may see more nvme slots. PCIe 3 lanes cost effectively nothing, they're just traces. 4.0 needs additional circuitry to amplify the signal past the first slot, I wouldn't expect it to be cheaper to add to a board ever. BangersInMyKnickers fucked around with this message at 02:29 on Mar 25, 2019 |
|
#
?
Mar 25, 2019 02:26
|
|
|
As has been said: 4.0 is more about making existing devices do what they currently do with fewer lanes, rather than expanding what existing devices are capable of doing. Had it come a year sooner, G-Sync might very well just run over the PCIe bus as well, instead of them porting NVlink back to SLI.
|
|
#
?
Mar 25, 2019 03:07
|
|
|
Unhappy Meal posted:Worse its more audiophile snake oil, an Elfidelity Power Filter Card. Filter your pci/pcie (it's double sided!) power for smooth noise free audio. They also make fan, sata, and usb versions. Still haven't found any indication it does anything since most people are doing multiple more sensible things in addition. If you want to talk about hardware that's decades behind in both drivers and hardware, look no further than audio interfaces. Khorne fucked around with this message at 03:21 on Mar 25, 2019 |
|
#
?
Mar 25, 2019 03:19
|
|
|
gepgepgep posted:I recently purchased parts to build a new system and seem to be encountering boot issues. The system won't post fully (get to BIOS announcement and won't respond to DEL or F2) and I'm getting a number of Q code errors. It seems to have more issues when either 1 or both sticks of RAM are in and in wondering if 1) is this a RAM Ryzen issue and/or 2) is there anything I can do to resolve? I had read on Reddit of someone using this exact RAM after changing settings...but I dont know how to without getting into BIOS. 1) Exchanged Mobo and CPU for new models, as the RAM was actually functional in the end. 2) Encountered a new error further into POST that was ultimately caused by a (cheap) DisplayPort cable. Switching to HDMI until I could order new DP 1.4 cables rectified this immediately. Thanks to all of the people who pointed out suggestions...especially those who made mention of looking at the final POST code vs. all the ones popping up.
|
|
#
?
Mar 25, 2019 15:31
|
|
|
LRADIKAL posted:https://cpu.userbenchmark.com/Compare/AMD-FX-8350-vs-AMD-Athlon-200GE/1489vsm592714 I kind of want to see a throw down between an FX-8370E and 200GE in their best configurations. I bet the FX still wins based on the testing I've seen, but the 200GE is only behind by 5-8% @ 3.9Ghz while consuming a quarter of the power.
|
|
#
?
Mar 25, 2019 18:09
|
|
|
Not AMD only, but this is pretty wild https://www.reddit.com/r/hardware/comments/b5ckcc/hackers_hijacked_asus_software_updates_to_install/
|
|
#
?
Mar 25, 2019 23:05
|
|
|
3peat posted:Not AMD only, but this is pretty wild https://www.reddit.com/r/hardware/comments/b5ckcc/hackers_hijacked_asus_software_updates_to_install/ Oof
|
|
#
?
Mar 25, 2019 23:35
|
|
|
I just got a 2600X and noticed a weird thing going on with CoreTemp/the cpu. When idling the temp goes from 45 down to 35, then spikes up to 45 and back down, and just loops. I thought the same thing was happening with the load/clocks in core temp but they don't seem to be related. This is from afterburner: 
|
|
#
?
Mar 25, 2019 23:54
|
|
|
Common pattern, use hysteresis or set your curves so you don't have fans spinning up and down constantly. This is also something water cooling does a good job with. Nothing to worry about in any case.
|
|
#
?
Mar 25, 2019 23:57
|
|
|
Heliosicle posted:I just got a 2600X and noticed a weird thing going on with CoreTemp/the cpu. When idling the temp goes from 45 down to 35, then spikes up to 45 and back down, and just loops. I thought the same thing was happening with the load/clocks in core temp but they don't seem to be related. I think that's normal with how XFR works on X-series chips. You're going to have a core that'll get boosted, request voltage for said boost and increase temps appropriately. XFR can be kinda bouncy on the desktop, my old 1700X and my current 2700X look the same in regards to temps.
|
|
#
?
Mar 25, 2019 23:58
|
|
|
Heliosicle posted:I just got a 2600X and noticed a weird thing going on with CoreTemp/the cpu. When idling the temp goes from 45 down to 35, then spikes up to 45 and back down, and just loops. I thought the same thing was happening with the load/clocks in core temp but they don't seem to be related. Then there's all the other obnoxious cooling to be done with fan curves and whatnot. BIOS updates can also make a big difference, sometimes for better, sometimes for worse.
|
|
#
?
Mar 26, 2019 00:21
|
|
|
Marketing is marketing, and this is a marketing piece. But it's still nice to see someone talking about actually doing something with all those cores, instead of just complaining about how a video game is 8FPS slower on a core--per-core basis. https://www.youtube.com/watch?v=UOJGP4NPM-g
|
|
#
?
Mar 26, 2019 07:29
|
|
|
taqueso posted:The board with more caps literally has some of each major type. It's like they went to the digikey cap section and picked cool looking versions of each.  I think they consider that a feature lol https://kitsunehifi.com/product/elfidelity-axf-104-ultra-power-supply-filter/
|
|
#
?
Mar 26, 2019 18:25
|
|
|
Munkeymon posted:
I'm willing to bet any amount of money that the caps on that card aren't even hooked up to anything.
|
|
#
?
Mar 27, 2019 01:33
|
|
|
No chokes allowed
|
|
#
?
Mar 27, 2019 07:52
|
|
|
mis-post, not that it will improve reception anyK8.0 posted:Adored, but officially on the books for Intel. Even adored is doing early damage control about how "sources tell me 7nm clocks aren't as high as expected" and how "4.5-4.7 might be more realistic". Oh no, just like sources should have told him. Also, that the Computex release was more realistic than CES. Maybe there will be another "last-minute readjustment". https://www.youtube.com/watch?v=9ioGZmorL8c&t=1557s (answer to adored mystery: AMD simply wanted to demonstrate beating the 9900K at much lower power, in a benchmark they already win even on Zen+, and not have people say "oh of course they would, it's twice the cores") https://www.youtube.com/watch?v=9ioGZmorL8c&t=1632s Now he's talking like it might be more efficient at lower clocks, but not clock super higher than previous nodes, as if this is just some node shrink like any other, rather than the second coming of ASML jesus. Boy, who could have predicted that. Maybe anyone who looked at the 22nm to 14nm move, or literally any other node shrink? https://www.youtube.com/watch?v=9ioGZmorL8c&t=1796s one very reliable source has told me this... "[motherboard partners] have had CPUs for some time, they've been at nowhere near the kind of clockspeeds you've mentioned.... and I've heard that Navi has had problems clocking at higher speeds than Radeon VII." https://www.youtube.com/watch?v=9ioGZmorL8c&t=1875s (also, as I said, let's pretend that you cram the same TDP into half the space... what happens? same thing as happened to Intel... die shrinks are not magic, if you clock them to hell and back then you don't get power savings, thermal density on something like coffee lake or comet lake (or matisse) is a critical problem) ... AMD put their best foot forward in that demo, the 2700X already beats the 9900K in Cinebench 15 MT IPC. If they made any IPC gains, they were clocked significantly under Intel. They will still be doing well to match Intel, according to The Stilt (~8% general deficit from Zen+ to Coffee Lake, excluding AVX2/AVX512 benchmarks). They need to make an 8% IPC gain and then also match Intel in clocks. Clocks don't scale perfectly either. I'm expecting that "you waited 2 years and spent $300 on an early access product that is now valued at $50 on the used market and then spent $400 on zen2 to get 95% of Intel single-thread performance" prediction I made 8 months ago to pan out overall. Intel user who spent $350 on a 7700K two years ago and sells it today gets... how much? Oh, right, full price. The 12/16C and especially Threadripper are still going to dumpster Intel, as predicted, but the whole "gamers are going to regret Coffee Lake" thing? Not really, probably. I think AMD and TSMC told the whole story when they said they got either twice the performance per watt... or 25% higher clocks... at the efficiency point. Not both, and not at the top of the curve. And cache coherency across the dies (i.e. not what AMD demonstrated) will probably sap performance a bit above that. I fully believe high-clocked 95W 8C parts... just not sure about the whole 16C 5.1 GHz at 120W thing. At least one of those things is a lie. Also, for the record: the IO die size is exactly what you get when you take a Zeppelin die (212 mm^2) and subtract two CCXs (44mm^2)... 122mm^2 for the new IO die, so actually a bit smaller. The fantasized L4 cache on the IO die probably does not exist, at least on Matisse (it may on Rome, and that may affect performance of Rome benchmarks relative to Mastisse). Some people are speculating that AMD managed to cut out or shrink some components on the IO die to make more room for cache, which... isn't the point of the IO die that it's the components that don't shrink? Most realistic speculation I've heard is that AMD might move to Samsung's 14nm node, which has a higher SRAM density and AMD might be able to use that on GF, which GF has licensed. Not that that's a huge probability. Paul MaudDib fucked around with this message at 14:15 on Mar 27, 2019 |
|
#
?
Mar 27, 2019 12:21
|
|
|
I have an insider leak which says that AMD will actually reach 5 GHz but only after a year of refinement. it's an engineering sample that won't be released to you guys for another year, don't wear it out. (note: I don't actually have a leak but it's just common sense and adored will probably claim that "his leak is valid over time" or some poo poo) Paul MaudDib fucked around with this message at 14:39 on Mar 27, 2019 |
|
#
?
Mar 27, 2019 13:58
|
|
|
do you think x399 will work on threadripper 3?
Paul MaudDib fucked around with this message at 15:02 on Mar 27, 2019 |
|
#
?
Mar 27, 2019 14:42
|
|
|
Are you ok?
|
|
#
?
Mar 27, 2019 14:59
|
|
|
yeah, just not a big follower of the whole "zen2 5.1 ghz" thing, as I've said right from the start. regardless, imagine 32 cores with full AVX2 performance, even at like 4.5 GHz  like, that should probably be a marketing campaign of some sort "what would you do with threadripper" Paul MaudDib fucked around with this message at 15:07 on Mar 27, 2019 |
|
#
?
Mar 27, 2019 15:04
|
|
|
|
| # ? May 17, 2024 15:51 |
|
|
FYI  goons, Newegg.ca has the 1700 for $239, sale ends in a week
|
|
#
?
Mar 27, 2019 15:24
|
|
