|
In case anyone cares, my 3570k/970 scores 1326.
|
#
?
Dec 14, 2016 06:56
|
|
|
|
| # ? Jun 5, 2024 18:06 |
|
|
PerrineClostermann posted:...Everyone's score is higher than mine by a lot. Is it the resolution? Well, let's do a little math. Ultrawide 1080p is 33% more pixels than a regular 1080p, your scores and average FPS are about 2/3rds that of the others... seems legit.
|
|
#
?
Dec 14, 2016 09:24
|
|
|
penus penus penus posted:I missed the AMD thing. So basically... No real info huh GPU-wise they showed "VEGA" running Battlefront with no frame rate information. I mean, was it even worth showing anything of vega?
|
|
#
?
Dec 14, 2016 11:27
|
|
|
Riflen posted:GPU-wise they showed "VEGA" running Battlefront with no frame rate information. I mean, was it even worth showing anything of vega? "We promise it's not vaporware"
|
|
#
?
Dec 14, 2016 14:20
|
|
|
DrDork posted:"We promise it's not vaporware" -bitboys
|
|
#
?
Dec 14, 2016 14:23
|
|
|
Riflen posted:GPU-wise they showed "VEGA" running Battlefront with no frame rate information. I mean, was it even worth showing anything of vega? No, it was a waste of time, pretty much other than to say "it's still coming" Edit: Here's a coincidence for you, unrelated to Vega, but related to the "my card wasn't seated properly so it was running at PCIe 1×" palaver earlier in the thread, TechPowerUp has just put out an article on the very subject of PCIe scaling, with a GTX 1080. HalloKitty fucked around with this message at 14:43 on Dec 14, 2016 |
|
#
?
Dec 14, 2016 14:34
|
|
|
HalloKitty posted:No, it was a waste of time, pretty much other than to say "it's still coming" What I am seeing is that as resolution increases, PCI-E lane width becomes less of a factor. In some instances, you actually be increased performance from PCI-E 3.0 8x over 3.0 16x.
|
|
#
?
Dec 14, 2016 14:51
|
|
|
Higher resolution = longer frame times = less bandwidth required between the gpu and the rest of your pc, since the things they talk about are very similar at any resolution. As for the increased performance of 8x over 16x, it's well within the margin of error, so it's not something I'd read too much into.
|
|
#
?
Dec 14, 2016 15:05
|
|
|
I hope vega ends up being good which then forces nvidia to cut prices which then causes a huge boom in nvidia sales and then people will be screaming to sue nvidia for anti competitive practices again
|
|
#
?
Dec 14, 2016 16:58
|
|
|
I hope vega is good so there's a boom in amahahahahahahahaha nope
|
|
#
?
Dec 14, 2016 17:06
|
|
|
I recently bought a PCIE x1 to x16 powered extender because I still had my 750 Ti laying around doing nothing, so when it arrived I thought I could use it as a dedicated PhysX card or a miner. When I plugged it in however...I got nothing, I plugged in the 4 pin slot for the PSU and it seemed to be working then, but my mobo and my Nvidia control panel still didn't recognise it, so I switched it with a wifi network card to see if it could pick up that...still nothing. I don't know whether my extender was faulty (I've read some reviews online that they got either working or faulty ones too) or my x1 slot just didn't recognise it. It's a shame really, I would've loved a multi-card configuration. Did anyone else have this problem?
|
|
#
?
Dec 14, 2016 17:06
|
|
|
Junior Jr. posted:I recently bought a PCIE x1 to x16 powered extender because I still had my 750 Ti laying around doing nothing, so when it arrived I thought I could use it as a dedicated PhysX card or a miner. When I plugged it in however...I got nothing, I plugged in the 4 pin slot for the PSU and it seemed to be working then, but my mobo and my Nvidia control panel still didn't recognise it, so I switched it with a wifi network card to see if it could pick up that...still nothing. From what I understand if you didn't pay some pretty big bux for that extender the chances of it working are very slim
|
|
#
?
Dec 14, 2016 17:08
|
|
|
penus penus penus posted:From what I understand if you didn't pay some pretty big bux for that extender the chances of it working are very slim This pretty much, PCIe wasn't really designed to be extended. The only one that is really reliable is a 3M one that costs a shitload, as you said.
|
|
#
?
Dec 14, 2016 17:15
|
|
|
HalloKitty posted:This pretty much, PCIe wasn't really designed to be extended. The only one that is really reliable is a 3M one that costs a shitload, as you said. Though to be fair I looked up whatever bulk pricing figures I could find and there is a ridiculous margin on them. Like even as a consumer if you were willing to buy 100 of them from a simple reseller you could cut the price by like half of the going rate. I'm sure they are spendy to make right but... not that spendy. Protip for anybody who wants to resell on amazon 
|
|
#
?
Dec 14, 2016 17:18
|
|
|
Well, to get into it on a more technical level, PCIe wasn't meant to be extended and suffers crosstalk, but the ribbon extension is also prone to voltage drop which is probably why it won't boot. There's a type of cheap riser meant to get around it called a powered riser, here's a few variations: Suplemental power from SATA or Molex, over USB conduit: http://www.ebay.com/itm/PCI-Express-Extension-Cable-1x-to-16x-with-Powered-Riser-/262348387316 Supplemental power from SATA over ribbon cable: http://www.ebay.com/itm/PCI-E-USB3-0-Express-1x-to16x-Extender-Riser-Board-Card-Adapter-SATA-Power-Cable-/311668600722 Supplemental power from Molex over ribbon cable: http://www.ebay.com/itm/PCI-E-1X-to-16X-Riser-Powered-Ribbon-Extender-Cable-30-CM-w-Molex-from-US-/262721828275 16x Molex Riser: http://www.ebay.com/itm/PCI-Express-PCIe-16X-16X-Riser-Card-Ribbon-Extend-Cable-Molex-Power-/162298968838 It's a buyer's market, since people mass produced these to cash in on bitcoin mining and now there's no demand and tons of supply. Edit: I noticed you said your riser was powered... make sure the power cable going to it is solid, check underneath the electrical tape. Some of them have a pretty shoddy soldering job on them and will break from fatigue. Zero VGS fucked around with this message at 17:47 on Dec 14, 2016 |
|
#
?
Dec 14, 2016 17:45
|
|
|
Junior Jr. posted:I recently bought a PCIE x1 to x16 powered extender because I still had my 750 Ti laying around doing nothing, so when it arrived I thought I could use it as a dedicated PhysX card or a miner. When I plugged it in however...I got nothing, I plugged in the 4 pin slot for the PSU and it seemed to be working then, but my mobo and my Nvidia control panel still didn't recognise it, so I switched it with a wifi network card to see if it could pick up that...still nothing. Also check that the 1X slot isn't disabled due to something in the Bios being turned on or another slot is filled killing the 1X from being used (check the manual for this really.) A lot of boards have weird things like this going on with some of the 1X slots.
|
|
#
?
Dec 14, 2016 18:44
|
|
|
SlayVus posted:What I am seeing is that as resolution increases, PCI-E lane width becomes less of a factor. In some instances, you actually be increased performance from PCI-E 3.0 8x over 3.0 16x. Truga posted:Higher resolution = longer frame times = less bandwidth required between the gpu and the rest of your pc, since the things they talk about are very similar at any resolution. Yeah, transferring models and textures and stuff is a one-time expense, everything needs to be in VRAM for the card to work well. So what's getting transferred during gameplay is actually mostly game-state updates and API commands. The faster your framerate, the more of those need to be sent per second, and lower resolutions allow faster framerates, ergo are more easily bottlenecked by PCIe transfer rates. 3.0x8 being faster than 3.016 is a bunch of bullshit though, what you're seeing is that they're essentially the same and the other sources of performance variability in the system are overwhelming any variation between the two.
|
|
#
?
Dec 14, 2016 20:57
|
|
|
EVGA RMA process went without a hitch. Got a new GTX1070 SC. Still coil whine. The next option would be somehow figuring out if its the MSI motherboard or the EVGA PSU, but i'm going to take option #2 and learn to deal with it.
|
|
#
?
Dec 15, 2016 06:32
|
|
|
Dali Parton posted:EVGA RMA process went without a hitch. Got a new GTX1070 SC. Still coil whine. The next option would be somehow figuring out if its the MSI motherboard or the EVGA PSU, but i'm going to take option #2 and learn to deal with it. That's a bummer. My old cheap 1440p korean monitor had coil whine and it sucked. I'm in the same boat though, my PSU started to have fan whine, which turned it from being nearly silent to being audible when idling. Normally I would oil the fan bearings but since it's built into the PSU I'm too scared to open it up
|
|
#
?
Dec 15, 2016 08:48
|
|
|
I just discovered that what I thought was insane GPU coil whine was just my speakers when they are turned off. Maybe check your peripherals? I used to have a display with maddening whine as soon it went into standby
|
|
#
?
Dec 15, 2016 08:55
|
|
|
oops wrong thread
|
|
#
?
Dec 15, 2016 09:15
|
|
|
My 1070 has coil whine too, but I can't hear it when it's closed up inside my case. Those fractal cases with the bitumen panels are legit.
|
|
#
?
Dec 15, 2016 11:01
|
|
|
I get more coil whine from my overhead LED lights than my computer. 
|
|
#
?
Dec 15, 2016 11:13
|
|
|
My 980Ti goes absolutely bonkers with coil whine in certain situations.
|
|
#
?
Dec 15, 2016 11:21
|
|
|
I haven't heard any coil whine from my 290X, and I have to open the side of the case to hear the very faint coil whine from my old 7950. It's never been a problem for me personally, not that that's a very interesting data point. vv Oh, not denying that, just saying I've been relatively lucky. I do know someone who bought a 7850 that whined incredibly loudly, it was so loud it was comical HalloKitty fucked around with this message at 15:05 on Dec 15, 2016 |
|
#
?
Dec 15, 2016 11:35
|
|
|
I've never had a GPU without at least some coil whine, it's absurdly common.
|
|
#
?
Dec 15, 2016 11:39
|
|
|
Haven't heard any coil whine from my Palit 1080, which is good cos the entire box is passive most of the time (and coils whining then would be a pita).
|
|
#
?
Dec 15, 2016 12:02
|
|
|
My EVGA 980Ti and 1080 both get coil whine. They both, however, only whine when not at idle and not under significant load. That generally means when I'm in the Battlefield 1 menus or in queue.
|
|
#
?
Dec 15, 2016 12:31
|
|
|
the real issue here is PC operator whine  Pretty much all my cards had coil whine, but i have a fractal define to muffle it. Sound dampening is so worth no side window
|
|
#
?
Dec 15, 2016 12:50
|
|
|
Most side windows that aren't tempered glass are this lovely soft acrylic that can't even absorb fingernails without marking anyways. Solid side panels that I can duct air from the outside directly to my video cards ftw!
|
|
#
?
Dec 15, 2016 13:20
|
|
|
Fauxtool posted:the real issue here is PC operator whine Haha maybe my card does have whine, got a nano s w/ no window Fractal make the best drat cases
|
|
#
?
Dec 15, 2016 14:57
|
|
|
AEMINAL posted:Fractal make the best drat cases Yes
|
|
#
?
Dec 15, 2016 15:06
|
|
|
Have coil whine? I'll bet when you look at your card's PCB you'll see some chokes that say "R22" printed on them. Those fuckers are in both AMD/Nvidia and are always the cause.
|
|
#
?
Dec 15, 2016 15:45
|
|
|
Yeah, I have some R22s. Good to know why it's there so I can't avoid it in the future!
|
|
#
?
Dec 15, 2016 16:12
|
|
|
Truga posted:Yeah, I have some R22s. Good to know why it's there so I can't avoid it in the future! There's no avoiding them, like literally everyone uses those loving things. If they just used the older version with the exposed coil, you'd be able to epoxy it. Card manufacturers hate us though.
|
|
#
?
Dec 15, 2016 16:19
|
|
|
Fauxtool posted:Pretty much all my cards had coil whine, but i have a fractal define to muffle it. Sound dampening is so worth no side window Never had coil whine on either of my 970s, but one of the WoW menus can cause it bad on my housemate's PC running the exact same model card. Luck of the draw I guess. Totally agreed on the case, with a windowless Define R5 my desktop sitting on my desk four feet from my head is inaudible over ambient noise when idle. When spun up to full it's still quiet enough to be muffled entirely by cheap Monoprice headphones.
|
|
#
?
Dec 15, 2016 16:48
|
|
Fauxtool posted:the real issue here is PC operator whine The plastic used in case windows has about the same muffing effects as a dampened case panel. Source: Fractal Design engineering & testing department.
|
|
|
#
?
Dec 15, 2016 16:53
|
|
|
wolrah posted:Never had coil whine on either of my 970s, but one of the WoW menus can cause it bad on my housemate's PC running the exact same model card. Luck of the draw I guess. This also has to do with whether you have VSync and/or frame limiting turned on. If you use neither, the GPU will try to render 200fps+ in the game menus and that will almost certainly produce whine.
|
|
#
?
Dec 15, 2016 16:54
|
|
|
AVeryLargeRadish posted:The plastic used in case windows has about the same muffing effects as a dampened case panel. Source: Fractal Design engineering & testing department. NZXT h440 has a window too, and it's meant to be about as quiet (with a similar system in) as the nice Fractal cases. I'd always thought most of the noise in a computer was due to noisy intake/exhaust fans, which can't really be dampened by side panels. The included Fractal G14 fans are pretty quiet, but switching them for something quieter would have a bigger effect than the makeup of the side panel. Coil whine chat: has anyone actually done the 'hot glue' fix?
|
|
#
?
Dec 15, 2016 17:14
|
|
|
|
| # ? Jun 5, 2024 18:06 |
|
|
I have no idea why they can't/don't use chokes with the coils glued. Though I've hit somewhat of a coil whine lottery in the past my present card has somewhat audible coil whine. Its not bad or anything but its a completely fanless system with no HDD so its kind of annoying to hear it when it *could* be silent
|
|
#
?
Dec 15, 2016 17:23
|
|