Urieldagda
Superclocked Member
- Total Posts : 113
- Reward points : 0
- Joined: 2008/07/23 11:06:50
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:43:09
(permalink)
EVGANewMember2719
2x MLCC is Nvidia spec. So I'd be happy with that. Probably wont get a stable overclock as high as a GPU with 6x MLCC though.
Maybe, maybe not. I have a feeling that there is a diminishing returns on these. Having 2 might be the best bang for the buck, more may not do very much. Oddly enough the FTW3 that Gamer's Nexus used had 6 POSCAPs while the Asus one had 6 MLCC and the FTW3 still had the best out of the box OC.. I think just having a few MLCC would just take interference out of the equation, heat and power will be more of an issue. BUT, liquid cooled might make a difference there. Maybe Gamer's Nexus should have tried the Strix under LN2, too!
|
EVGANewMember2719
New Member
- Total Posts : 71
- Reward points : 0
- Joined: 2015/11/18 01:31:24
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:43:17
(permalink)
LordNVader
gpkgpk
Worth repeating EVGA FTW3 pics https://imgur.com/a/IMVFdTP (Scroll down to 2nd pic)
Thats nice to know. So 2 shouldn't have problems since thats nvdias specs right
Ya it will operate just as good as the founder edition. But you probably can't go as high as a Strix card with 6x MLCC. That is why Strix OC has advertised the highest out of any vendor. 1900+mhz.
|
Desaccorde
Superclocked Member
- Total Posts : 104
- Reward points : 0
- Joined: 2014/11/29 10:42:43
- Status: offline
- Ribbons : 1
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:47:19
(permalink)
Urieldagda
EVGANewMember2719
2x MLCC is Nvidia spec. So I'd be happy with that. Probably wont get a stable overclock as high as a GPU with 6x MLCC though.
Maybe, maybe not. I have a feeling that there is a diminishing returns on these. Having 2 might be the best bang for the buck, more may not do very much. Oddly enough the FTW3 that Gamer's Nexus used had 6 POSCAPs while the Asus one had 6 MLCC and the FTW3 still had the best out of the box OC.. I think just having a few MLCC would just take interference out of the equation, heat and power will be more of an issue.
BUT, liquid cooled might make a difference there. Maybe Gamer's Nexus should have tried the Strix under LN2, too!
That's because of the Power Limit difference. You should compare FTW3 Ultra to Asus Strix OC because they're on the same segment. TUF was not a match for FTW3 anyway, so it was like apples and oranges.
|
Lokator
Superclocked Member
- Total Posts : 147
- Reward points : 0
- Joined: 2018/03/13 18:25:57
- Status: offline
- Ribbons : 1
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:47:26
(permalink)
MLCC and POSCAP might be similar pricy to buy, but POSCAP might be easier to solder on the board ---> less work time ---> cheaper (yes, I am a noob) - Asus heavily advertises its fully automated board production process, so for them it might not be an issue to solder more of these smaller MLCC on these boards? - fully speculated from a noob oh and nice, Jayz made a video about it, will blow up now I guess
post edited by Lokator - 2020/09/25 14:50:04
|
robertdinh
Superclocked Member
- Total Posts : 139
- Reward points : 0
- Joined: 2020/08/10 09:42:44
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:48:40
(permalink)
EVGANewMember2719
LordNVader
gpkgpk
Worth repeating EVGA FTW3 pics https://imgur.com/a/IMVFdTP (Scroll down to 2nd pic)
Thats nice to know. So 2 shouldn't have problems since thats nvdias specs right
Ya it will operate just as good as the founder edition. But you probably can't go as high as a Strix card with 6x MLCC. That is why Strix OC has advertised the highest out of any vendor. 1900+mhz.
Strix looking kind of tempting not gonna lie
|
tidus316
New Member
- Total Posts : 14
- Reward points : 0
- Joined: 2009/12/22 23:15:25
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:51:34
(permalink)
Is EVGA going to make an announcement about this? If not for once I'll have to go with ASUS. I really don't want to do that.
|
degamad
New Member
- Total Posts : 15
- Reward points : 0
- Joined: 2016/07/07 12:12:29
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:52:24
(permalink)
Anything official from EVGA on this? Founders is looking like a safer bet.
|
pixelcowboy
Superclocked Member
- Total Posts : 213
- Reward points : 0
- Joined: 2017/06/16 17:42:43
- Status: offline
- Ribbons : 1
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:53:02
(permalink)
I'm not getting crashes on my XC3, but I've noticed that boost clocks jump like crazy and I can't overclock for sh***. Well, I can, but I keep getting lower scores with 'stable' overclocks over the stock. Even raising power limit hurts performance.
|
YURIIII
New Member
- Total Posts : 48
- Reward points : 0
- Joined: 2016/09/20 09:41:07
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:56:23
(permalink)
pixelcowboy
I'm not getting crashes on my XC3, but I've noticed that boost clocks jump like crazy and I can't overclock for sh***. Well, I can, but I keep getting lower scores with 'stable' overclocks over the stock. Even raising power limit hurts performance.
Can you post a picture of capacitors in your XC3, are they are all POSCAPS?
|
Yesterdays_news
New Member
- Total Posts : 100
- Reward points : 0
- Joined: 2014/11/04 16:51:30
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:57:42
(permalink)
Could such a thing be fixed with driver or bios updates? this is troubling
|
robertdinh
Superclocked Member
- Total Posts : 139
- Reward points : 0
- Joined: 2020/08/10 09:42:44
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:57:43
(permalink)
stefanhondasi
I hope evga goes bankrupt 😉
I hope not, they are way more consumer-friendly than the other aibs. Especially in the west.
|
Gr3yGhost
Superclocked Member
- Total Posts : 230
- Reward points : 0
- Joined: 2015/05/16 09:52:05
- Status: offline
- Ribbons : 1
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:57:54
(permalink)
Lokator
Will EVGA change it if true?
I hope all the AIB's do so when it comes to creating new batches. It would delay creating new supply to fill the demand. I have no idea as to the complexity of the tooling needed to precisely install such small devices, but from what i've seen in some assembly videos - they use automation for the majority of production, they could conceivably re-program the automated machine(s) that specifically deals with installing these to add 1 or more additional groups of these capacitors. This does however, increase cost of materials, slows production for a period of time - and causes additional cost to replace anyone who would wish (likely anyone who has this concern) with an updated board. The trickle-down effect of making such a production change could be costly. And as someone who runs a business, you want to use the most cost-efficient method as much as possible while also trying to deliver the same quality product. Watching Jayz video, I agree with him that the "free" option is a bios update to lower the clocks. But I am crossing my fingers, that all AIB's make a change for future production runs of these cards especially on higher tier cards. I'll also cross my toes hoping that the 3080 Hybrid has these 2 groups of MLCCs at the very least
|
EVGANewMember2719
New Member
- Total Posts : 71
- Reward points : 0
- Joined: 2015/11/18 01:31:24
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 14:58:24
(permalink)
Yesterdays_news
Could such a thing be fixed with driver or bios updates? this is troubling
Ya they will release a bios that will lower the clocks. So worse performance.
|
robbazking
New Member
- Total Posts : 17
- Reward points : 0
- Joined: 2009/07/10 14:05:20
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:00:04
(permalink)
Only bought Gigabyte and EVGA card in the past, when it turned out the 3090 gigabyte gaming oc had 6 poscap's I probably sending it back.
Turns out EVGA hasn't done much better? So all in all the safe bets are ASUS and FE?
Rampage IV Gene - i7 4930k @ 4,7ghz - Gtx 780 Classified.
|
Desaccorde
Superclocked Member
- Total Posts : 104
- Reward points : 0
- Joined: 2014/11/29 10:42:43
- Status: offline
- Ribbons : 1
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:00:21
(permalink)
stefanhondasi
I hope evga goes bankrupt 😉
Chill mate.
|
repo1979
FTW Member
- Total Posts : 1116
- Reward points : 0
- Joined: 2017/02/21 20:37:56
- Location: UTAH
- Status: offline
- Ribbons : 4
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:00:39
(permalink)
EVGANewMember2719
Now we see why Asus Strix is considered the best card and not just "Paying for ROG" Branding. They got all 6 MLCC.
Looks like i will go with a Asus 3080 20gb Oc edition card as well then if this is the case after-all
|
EVGANewMember2719
New Member
- Total Posts : 71
- Reward points : 0
- Joined: 2015/11/18 01:31:24
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:02:57
(permalink)
repo1979
EVGANewMember2719
Now we see why Asus Strix is considered the best card and not just "Paying for ROG" Branding. They got all 6 MLCC.
Looks like i will go with a Asus 3080 20gb Oc edition card as well then if this is the case after-all
Ya, I backordered 3080 Strix OC and I have a FTW Ultra on the way already. I'll test both. I can sell the one I don't like so not really worried about returning it.
|
YURIIII
New Member
- Total Posts : 48
- Reward points : 0
- Joined: 2016/09/20 09:41:07
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:04:23
(permalink)
It feels like they are artificially limiting clockspeeds by adjusting amount of MLCC:s on their boards. If this rumor is true, feels pretty ****ty.
I understand if XC3's have 1 or 2 and somewhat limited boost clocks due that, because they are a bit cheaper but FTW3 with premium price and limited boosts, hell naw. Asus can promise +1900MHz with less money.
|
fragility_V1
SSC Member
- Total Posts : 940
- Reward points : 0
- Joined: 2015/07/16 10:02:01
- Location: Southern California
- Status: offline
- Ribbons : 4
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:11:38
(permalink)
Just opened my 3090 FTW3 has 4x POSCAP and 2x MLCC
|
Yesterdays_news
New Member
- Total Posts : 100
- Reward points : 0
- Joined: 2014/11/04 16:51:30
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:27:11
(permalink)
Hmm I’ll be fine with slightly slower performance I guess if that’s the end result of this. Never was one for over clocking
|
Drwaffles
New Member
- Total Posts : 95
- Reward points : 0
- Joined: 2015/12/29 14:57:48
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:29:07
(permalink)
|
Veratu
Superclocked Member
- Total Posts : 154
- Reward points : 0
- Joined: 2004/06/06 16:28:01
- Status: offline
- Ribbons : 1
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:29:30
(permalink)
If these cards are failing (eVGA cards) at normal shipped specs, I'd raise a question for cause, but so far everyone who has an eVGA card in this thread has said they're NOT having problems. Before you go on the witch hunt give the companies some time to diagnose the issue. Just because Igor discovered variances doesn't mean that's the underlying issue or that your card is of poor quality. Problems like this are not always easily identified in a matter of hours which is what you're asking. That's unrealistic, especially given the demand and whirlwind of other headaches all of these companies are faced with given the volumes / supply and the crappy launch in general. Sometimes issues like this can take days to isolate and prove out, even with the best test equipment in the world. If you're not an electrical engineer with intimate knowledge of the specific design of each AIBs card, then you too are speculating based on likely misinformation. 1 MLCC, 2 MLCC, 6 MLCC, none of that may be relevant as long as the components they did use meet the specifications of the design.
If you're complaining because you can't get an extra 200mhz by overclocking because of these components, then I have no sympathy. The card is performing at the specs it's rated at without problems, that's what you bought. If you have issues pushing it beyond those specs, that's on you, not the company. It's doing what it's advertised to do.
Until out of the box eVGA cards are not performing at out of the box specs, you really have no room to complain. Yes I know other companies are having "reported" issues, but without complete details on those issues that's all speculation at this point too. These GPUs are already ridiculously low yield at foundary specs that are hard to achieve in general. The fact that they work it all at these performance levels is amazing in of itself.
Everyone just needs to chill out, we're on day 2 of the 3090 launch and week 2 of the 3080 launch, there isn't enough data yet to make a conclusion and it's highly likely they will isolate the issue long before your card can be returned for being faulty, so give them the chance to figure it out, with AMD's launch around the corner I'm sure the desire to keep all of the customers happy will be high.
Running a products company is hard, especially when you have supply chain problems. It sounds like this whole launch had a mix of issues. Let the dust settle a bit before you go of the rails, they're all working their butts off right now to handle a slew of problems, most of which are not solved in minutes which is what your expectation is.
|
joeeoh
New Member
- Total Posts : 31
- Reward points : 0
- Joined: 2014/05/10 14:45:25
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:31:24
(permalink)
Veratu
If these cards are failing (eVGA cards) at normal shipped specs, I'd raise a question for cause, but so far everyone who has an eVGA card in this thread has said they're NOT having problems. Before you go on the witch hunt give the companies some time to diagnose the issue. Just because Igor discovered variances doesn't mean that's the underlying issue or that your card is of poor quality. Problems like this are not always easily identified in a matter of hours which is what you're asking. That's unrealistic, especially given the demand and whirlwind of other headaches all of these companies are faced with given the volumes / supply and the crappy launch in general. Sometimes issues like this can take days to isolate and prove out, even with the best test equipment in the world. If you're not an electrical engineer with intimate knowledge of the specific design of each AIBs card, then you too are speculating based on likely misinformation. 1 MLCC, 2 MLCC, 6 MLCC, none of that may be relevant as long as the components they did use meet the specifications of the design.
If you're complaining because you can't get an extra 200mhz by overclocking because of these components, then I have no sympathy. The card is performing at the specs it's rated at without problems, that's what you bought. If you have issues pushing it beyond those specs, that's on you, not the company. It's doing what it's advertised to do.
Until out of the box eVGA cards are not performing at out of the box specs, you really have no room to complain. Yes I know other companies are having "reported" issues, but without complete details on those issues that's all speculation at this point too. These GPUs are already ridiculously low yield at foundary specs that are hard to achieve in general. The fact that they work it all at these performance levels is amazing in of itself.
Everyone just needs to chill out, we're on day 2 of the 3090 launch and week 2 of the 3080 launch, there isn't enough data yet to make a conclusion and it's highly likely they will isolate the issue long before your card can be returned for being faulty, so give them the chance to figure it out, with AMD's launch around the corner I'm sure the desire to keep all of the customers happy will be high.
Running a products company is hard, especially when you have supply chain problems. It sounds like this whole launch had a mix of issues. Let the dust settle a bit before you go of the rails, they're all working their butts off right now to handle a slew of problems, most of which are not solved in minutes which is what your expectation is.
I've been crashing and freezing so much across every title I play that my machine straight shut off on me playing Rise of the Tomb Raider. I'm literally about to swap back out for my GTX 1080, unless there's a software update, or a replacement PSU gets delivered.
|
degenerate
SSC Member
- Total Posts : 614
- Reward points : 0
- Joined: 2013/08/23 05:38:31
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:34:59
(permalink)
fragility_V1
Just opened my 3090 FTW3 has 4x POSCAP and 2x MLCC
They should probably update their product page photos to reflect this then, because as of right now it is showing 6x POSCAP and 0x MLCC. Not a good look if there are actually 4x/2x.
 |5900x|MSI B550 Unify-X|2x16GB G.Skill 3800 C14 @ 1900 FCLK|2TB WD SN850 M.2|EVGA 3090 KPE|EVGA 1600T2|Alienware '25' & LG 48" CX OLED|
|
nolaes619
New Member
- Total Posts : 63
- Reward points : 0
- Joined: 2014/11/21 07:21:44
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:37:22
(permalink)
Current Daily Driver:
Intel 8700K (6.8)Asus ROG Strix 370-E GamingCorsair 16GB 3000MHz Vengeance LPXCaptain DeepCool 240EXEVGA GeForce FTW3 3080 ULTRASamsung 960EVO Series 250GB PCIe NVMe - M.2Samsung 860 EVO 1TB 2.5Corsair HXi Series HX1000iCorsair Graphite 760TLG CX55 120HZ/GYSNC Ready :)
|
Veratu
Superclocked Member
- Total Posts : 154
- Reward points : 0
- Joined: 2004/06/06 16:28:01
- Status: offline
- Ribbons : 1
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:37:27
(permalink)
joeeoh
Veratu
If these cards are failing (eVGA cards) at normal shipped specs, I'd raise a question for cause, but so far everyone who has an eVGA card in this thread has said they're NOT having problems. Before you go on the witch hunt give the companies some time to diagnose the issue. Just because Igor discovered variances doesn't mean that's the underlying issue or that your card is of poor quality. Problems like this are not always easily identified in a matter of hours which is what you're asking. That's unrealistic, especially given the demand and whirlwind of other headaches all of these companies are faced with given the volumes / supply and the crappy launch in general. Sometimes issues like this can take days to isolate and prove out, even with the best test equipment in the world. If you're not an electrical engineer with intimate knowledge of the specific design of each AIBs card, then you too are speculating based on likely misinformation. 1 MLCC, 2 MLCC, 6 MLCC, none of that may be relevant as long as the components they did use meet the specifications of the design.
If you're complaining because you can't get an extra 200mhz by overclocking because of these components, then I have no sympathy. The card is performing at the specs it's rated at without problems, that's what you bought. If you have issues pushing it beyond those specs, that's on you, not the company. It's doing what it's advertised to do.
Until out of the box eVGA cards are not performing at out of the box specs, you really have no room to complain. Yes I know other companies are having "reported" issues, but without complete details on those issues that's all speculation at this point too. These GPUs are already ridiculously low yield at foundary specs that are hard to achieve in general. The fact that they work it all at these performance levels is amazing in of itself.
Everyone just needs to chill out, we're on day 2 of the 3090 launch and week 2 of the 3080 launch, there isn't enough data yet to make a conclusion and it's highly likely they will isolate the issue long before your card can be returned for being faulty, so give them the chance to figure it out, with AMD's launch around the corner I'm sure the desire to keep all of the customers happy will be high.
Running a products company is hard, especially when you have supply chain problems. It sounds like this whole launch had a mix of issues. Let the dust settle a bit before you go of the rails, they're all working their butts off right now to handle a slew of problems, most of which are not solved in minutes which is what your expectation is.
I've been crashing and freezing so much across every title I play that my machine straight shut off on me playing Rise of the Tomb Raider. I'm literally about to swap back out for my GTX 1080, unless there's a software update, or a replacement PSU gets delivered.
Have you isolated your problem to the card? Do you have a watt meter to test your draw at the machine level with both cards? Have you tested your power supply? What card are you using? Is it overclocked or out of the box settings? I had a repeat crashing issue in a prior generation of cards that turned out to be a faulty component in my power supply that only showed up when the power supply heated up, which coincided with the card heating up and led me to believe it was the card. When I swapped cards the problem went away, because the other card drew less power and the component in the PSU didn't get as warm, thus it held up. You gotta do a little bit of leg work to try and narrow it down before everyone calls foul on the card. That doesn't mean your card isn't faulty, it very well could be, but be certain or highly certain before calling foul. Side note, that sucks you're crashing a lot, that's a crappy experience no matter how you look at it... hopefully you get it sorted out.
|
joeeoh
New Member
- Total Posts : 31
- Reward points : 0
- Joined: 2014/05/10 14:45:25
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:50:29
(permalink)
Veratu
Have you isolated your problem to the card? Do you have a watt meter to test your draw at the machine level with both cards? Have you tested your power supply? What card are you using? Is it overclocked or out of the box settings?
I had a repeat crashing issue in a prior generation of cards that turned out to be a faulty component in my power supply that only showed up when the power supply heated up, which coincided with the card heating up and led me to believe it was the card. When I swapped cards the problem went away, because the other card drew less power and the component in the PSU didn't get as warm, thus it held up. You gotta do a little bit of leg work to try and narrow it down before everyone calls foul on the card. That doesn't mean your card isn't faulty, it very well could be, but be certain or highly certain before calling foul.
Side note, that sucks you're crashing a lot, that's a crappy experience no matter how you look at it... hopefully you get it sorted out.
I'm about to narrow things with the GTX 1080 swap. I don't have a wattmeter from what I've researched anything that runs between PSU and wall outlet won't be accurate or sample the consumption quick enough to detect intersample-peaking. That would leave the Nvidia (PCAT?) as the only candidate for accuracy but they don't seem available to the public. I can't test a PSU swap until I get the replacement delivered going from 650W to 750W, never used GPU HW scheduling, I've run it both OC'd and stock with little to no difference in stability. In fact my longest crash-free sessions was where I used a +67 core offset and a +200 Mem offset recommended by the X1 OC scanner. The most confusing thing about the crashes is it's completely random how stable things have been throughout the week even with the same work loads. Sometimes it can't get off the ground sometimes I can get 5 hours of solid playtime on it but as soon as I exit and reload it's back to crashing. I'm sure EVGA will have my back whatever happens going forward and I'm also confident the availability of only a single driver is playing a factor. edit: I've continued this conversation in another thread due to topic relevance, here: https://forums.evga.com/FindPost/3095714
post edited by joeeoh - 2020/09/25 21:20:46
|
legit
New Member
- Total Posts : 68
- Reward points : 0
- Joined: 2005/06/19 06:32:21
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 15:59:04
(permalink)
If it was as simple as 6 MLCCs = Good, why are there quite a number of Asus TUF cards just crashing though?
|
jPup omg
New Member
- Total Posts : 90
- Reward points : 0
- Joined: 2020/07/09 23:22:13
- Location: Digging through a warehouse of empty 3080 boxes
- Status: offline
- Ribbons : 0
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 16:01:48
(permalink)
|
arestavo
CLASSIFIED ULTRA Member
- Total Posts : 6916
- Reward points : 0
- Joined: 2008/02/06 06:58:57
- Location: Through the Scary Door
- Status: offline
- Ribbons : 76
Re: Igors Lab: Less MLCC, more crashing and lower clocks?
2020/09/25 16:02:32
(permalink)
Jacob didn't address this at all during their livestream that just ended. I asked, and a whole bunch of other people asked.
If you have a Twitter account, maybe keep asking him until he answers.
|