[GamersNexus] Building a GTX 1080 Hybrid P 3: Performance Results (Overclocking 1080)

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Grooveriding

Diamond Member
Dec 25, 2008
9,108
1,260
126
nobody said the actual OC was bad.

1. question of 2.5Ghz claims
2. so-so scaling. or at least fps gains

The 2.5 was a junk rumour from videocardz as I recall. It's never going to happen looking at launch reviews. A card might hit that clock with LN2 and an additional PCB of power delivery soldered on, maybe even higher. The chance of seeing a board released with even a 2.2ghz clock is very low when the average clock these boards are doing is around 2ghz.

Even an AIB card can only go so far as to bin a better chip and increase the power limit in BIOS. This generally can net you a few 100mhz. After that you need voltage, so you need a BIOS mod or an EVGA Classified.

2.5ghz Videocardz was desperate for some more clicks.
 

LTC8K6

Lifer
Mar 10, 2004
28,520
1,575
126
Well, no one is forcing anyone to buy a FE card.

I don't understand why anyone would.

Unless the wait for better versions is going to be a lot longer than most people think?
 

flopper

Senior member
Dec 16, 2005
739
19
76
The 2.5 was a junk rumour from videocardz as I recall. It's never going to happen looking at launch reviews. A card might hit that clock with LN2 and an additional PCB of power delivery soldered on, maybe even higher. The chance of seeing a board released with even a 2.2ghz clock is very low when the average clock these boards are doing is around 2ghz.

Even an AIB card can only go so far as to bin a better chip and increase the power limit in BIOS. This generally can net you a few 100mhz. After that you need voltage, so you need a BIOS mod or an EVGA Classified.

2.5ghz Videocardz was desperate for some more clicks.

2ghz and maybe 50 hz extra is what we are seeing with todays FE cards.
thirdparty might do more but then we dont really know how they scale on finfet yet.
 

Det0x

Golden Member
Sep 11, 2014
1,065
3,116
136
Guess i can copy my post over to this thread also, as its relevant to the stock cooler performance.

980 Ti and Titan X have 601 mm² to transfer the heat/energy, while 1080 only have 314 mm².

250 W / 601 mm² = ~0.415 W per mm² (stock)
350 W / 601 mm² = ~0.582 W per mm² (max overclock)

Or in the case of the heatpipe based 980

165 W / 398 mm² = ~0.414 W per mm² (stock)
215 W / 398 mm² = ~0.540 W per mm² (max overclock)

Founders 1080

180 W / 314 mm² = ~0.573 W per mm² (stock)
230 W / 314 mm² = ~0.732 W per mm² (max overclock)
 
Last edited:

sirmo

Golden Member
Oct 10, 2011
1,014
391
136
Highly doubt the 2.5Ghz has any truth to it. To be fair 2.1Ghz is pretty high as it is. I doubt you'll get past 2.2 without LN2, or some golden samples.
 

Timmah!

Golden Member
Jul 24, 2010
1,463
729
136
Highly doubt the 2.5Ghz has any truth to it. To be fair 2.1Ghz is pretty high as it is. I doubt you'll get past 2.2 without LN2, or some golden samples.

TBH, i have a feeling that 2,25 - 2,3 GHz will be about the ceiling what the aftermarket cards with another 6-pin/8-pin will do. I mean, no custom BIOS or LN2, no modding of this kind considered.

I would wish for 2,4-2,5 GHz OC to be true, but if the past is any indicator of what future might hold, i dont see it. But you never know.
 

Flapdrol1337

Golden Member
May 21, 2014
1,677
93
91
Guess i can copy my post over to this thread also, as its relevant to the stock cooler performance.

980 Ti and Titan X have 601 mm² to transfer the heat/energy, while 1080 only have 314 mm².

250 W / 601 mm² = ~0.415 W per mm² (stock)
350 W / 601 mm² = ~0.582 W per mm² (max overclock)

Or in the case of the heatpipe based 980

165 W / 398 mm² = ~0.414 W per mm² (stock)
215 W / 398 mm² = ~0.540 W per mm² (max overclock)

Founders 1080

180 W / 314 mm² = ~0.573 W per mm² (stock)
230 W / 314 mm² = ~0.732 W per mm² (max overclock)
So? the 6700K overclocks just fine and does 91W / 122mm²=0.74 W per mm² at stock, and half of that is gpu, cpu part is probably way above. Also has an extra barrier with crappy paste and metal lid.

What the 1080 will clock to is more down to how much volts nvidia thinks is safe.
 
Mar 10, 2006
11,715
2,012
126
TBH, i have a feeling that 2,25 - 2,3 GHz will be about the ceiling what the aftermarket cards with another 6-pin/8-pin will do. I mean, no custom BIOS or LN2, no modding of this kind considered.

I would wish for 2,4-2,5 GHz OC to be true, but if the past is any indicator of what future might hold, i dont see it. But you never know.

I think 2.2-2.3GHz is a reasonable limit to expect for the aftermarket stuff. Still a very good overclock if true, though. We'll find out soon.
 

sirmo

Golden Member
Oct 10, 2011
1,014
391
136
TBH, i have a feeling that 2,25 - 2,3 GHz will be about the ceiling what the aftermarket cards with another 6-pin/8-pin will do. I mean, no custom BIOS or LN2, no modding of this kind considered.

I would wish for 2,4-2,5 GHz OC to be true, but if the past is any indicator of what future might hold, i dont see it. But you never know.
The gamersnexus guy downclocked and undervolted the RAM to divert all the power to the GPU core, and he wasn't able to get close to 2.2Ghz.

Even 2.1Ghz wasn't sustainable. Perhaps with some better batches in the future. But I don't think it will be common.
 
Mar 10, 2006
11,715
2,012
126
The gamersnexus guy downclocked and undervolted the RAM to divert all the power to the GPU core, and he wasn't able to get close to 2.2Ghz.

Even 2.1Ghz wasn't sustainable. Perhaps with some better batches in the future. But I don't think it will be common.

How much power do you think the GDDR5X needs? Anyway, we'll see.
 

sirmo

Golden Member
Oct 10, 2011
1,014
391
136
How much power do you think the GDDR5X needs? Anyway, we'll see.

Tried to figure it out but Micron's datasheet doesn't list current draw. It is not insignificant though would be my guess. One of the reasons HBM is a big deal is because HBM memory is clocked lower which saves quite a bit of power.
 

kawi6rr

Senior member
Oct 17, 2013
567
156
116
Interesting video. NVIDIA isn't being very clear on why the Founders Edition is $100 more. It seems like they want to test the market to see if they can sell it as more of A premium product.

Either way all the sheep out there are going to buy this thing as quick as they can and Nvidia will make tons of money off them.


Threadcrapping and trolling are not allowed
Markfw900
 
Last edited by a moderator:

tviceman

Diamond Member
Mar 25, 2008
6,734
514
126
www.facebook.com
I think 2.2-2.3GHz is a reasonable limit to expect for the aftermarket stuff. Still a very good overclock if true, though. We'll find out soon.

I don't think there will be a single Pascal card able to hit 2.3ghz without bios power limit and voltage mods. I'm confident in that prediction. 2.2ghz is going to be the upper end, with rare "ungodly" 2.23ghz validations.
 
Last edited:

Actaeon

Diamond Member
Dec 28, 2000
8,657
20
76
2.1Ghz is going to be the realistic overclock for these. This is what Nvidia demo'ed at the reveal. If they could do 2.5ghz, there is no way they would not have demoed it at just 2.1ghz. If they had chips that would go higher they would have demoed that instead.

This 2.1ghz barrier also matches up with the reviews we are seeing too.
 

Lepton87

Platinum Member
Jul 28, 2009
2,544
9
81
I have a suspicion that even those review samples are cherry-picked and retail samples will usually overclock lower.
 

DDH

Member
May 30, 2015
168
168
111
As much as I like to see people holding Nvidia culpable for their marketing claims.... its getting a bit overboard.

Clearly the card is maxing out the available power draw of the single 8 pin. Once a custom AIB board with greater power delivery is available then Im sure these cards will break 2.2 regularly, maybe 2.3.
 

Elixer

Lifer
May 7, 2002
10,376
762
126
Well, no one is forcing anyone to buy a FE card.

I don't understand why anyone would.

Unless the wait for better versions is going to be a lot longer than most people think?

Maybe they will give them a bone and include some kind of a game coupon in it.
Otherwise, yeah, makes no sense at all now.
 

Abwx

Lifer
Apr 2, 2011
11,172
3,869
136
As much as I like to see people holding Nvidia culpable for their marketing claims.... its getting a bit overboard.

Clearly the card is maxing out the available power draw of the single 8 pin. Once a custom AIB board with greater power delivery is available then Im sure these cards will break 2.2 regularly, maybe 2.3.

A switch mode PSU that deliver 1V will not be impacted by a 12V rail that sink to say 10V, so it s not a question of connector, besides this wouldnt retain the card from booting at 2.2-2.3 since it s not loaded and that the existing connector is hence overkill for this situation..
 

dark zero

Platinum Member
Jun 2, 2015
2,655
138
106
Ok.. I was expecting that the OCed cards will some sooner... Sadly that is not the case... However... I wonder what will their limits... Maybe a GPU as fast as Intel Processor?
 

R0H1T

Platinum Member
Jan 12, 2013
2,582
162
106
I don't think there will be a single Pascal card able to hit 2.3ghz without bios power limit and voltage mods. I'm confident in that prediction. 2.2ghz is going to be the upper end, with rare "ungodly" 2.23ghz validations.
Definitely, also this ~
2.1Ghz is going to be the realistic overclock for these. This is what Nvidia demo'ed at the reveal. If they could do 2.5ghz, there is no way they would not have demoed it at just 2.1ghz. If they had chips that would go higher they would have demoed that instead.

This 2.1ghz barrier also matches up with the reviews we are seeing too.
For anything above 2.1~2.2GHz there's a good chance an additional power connector will be required, that & possibly water cooling as well. Only golden samples would be exception to the norm, the number of reviews not being able to hit 2.1GHz is telling.
 

Slaughterem

Member
Mar 21, 2016
77
23
51
I guess I could copy my post from a different forum to here.

Is 2300 mhz a pipe dream? Lets just take a look at TSCM 16FF+ technology.

http://www.tsmc.com/english/dedicatedFoundry/technology/16nm.htm
Quote:
TSMC's 16FF+ (FinFET Plus) technology can provide above 65 percent higher speed, around 2 times the density, or 70 percent less power than its 28HPM technology. Comparing with 20SoC technology, 16FF+ provides extra 40% higher speed and 60% power saving. By leveraging the experience of 20SoC technology, TSMC 16FF+ shares the same metal backend process in order to quickly improve yield and demonstrate process maturity for time-to-market value.

Now remember this is being applied to the 980 based process. Looking at the power of the 1080 it appears Nvidia went only with performance and no power savings. You can see this in the charts from Tech power up looking at 1080 and 980 typical game consumption.
https://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_1080/24.html

Since they claim above 65% higher speed and we see that they did not try for reduced power lets use 70%. A typical 980 has a base clock of 1126 mhz with an ASUS strix card having a 1240 mhz clock.
http://www.computerbase.de/2016-05/grafikkarten-testsystem-2016/2/

So lets do the math 70% of 1240 is 868 mhz which equals 2104 for max frequency. at same power as a ASUS Strix which is 190 w. The Strix can OC to 1481 so add 168 more to 2104 and you get 2272 mhz. This is a best case scenario with TDP moving into the > than 200 watt area. The problem then becomes temperature of the chip and would cause the card to throttle just as we are seeing with the reference 1080 only averaging a 1778 mhz clock frequency in games after 20 mins. Nvidia showed us a clock of 2141 mhz during their launch, do you think if they could have gone a lot higher they would not have displayed that capability?

http://www.computerbase.de/2016-05/geforce-gtx-1080-test/6/
Quote:
The maximum achievable 1,886 MHz, the GeForce GTX 1080 Founders Edition are more of a theoretical value: In games you never see this clock. There, the highest ever measured frequency was during the tests 1,835 MHz, but was also the clock after a few seconds history. In demanding titles 1.785 MHz has finally placed as the highest, realistic clock out.
I would hope that now since everyone has had a day to become more objective the hype train would slow down. Don't misunderstand my point of this card, it is a good product and is clear that Nvidia went for max performance. The problem I have is a 980TI is only 13% slower and the price of a 1080 ASUS Strix will not be less than $700.
 

Mahigan

Senior member
Aug 22, 2015
573
0
0
They're trying to make money before AMD strikes. That's why the release is filled with marketing and hype while the product isn't really that impressive imo.

Charging the $100 premium before AMD launches Polaris. Polaris won't be competing with the likes of a GTX 1080 and maybe not even a GTX 1070 but the price is likely to be quite attractive as well as the DX12 performance.
 

thesmokingman

Platinum Member
May 6, 2010
2,307
231
106
Something else to consider is that the samples sent out for review were mostly likely their highest binned parts from their available stock. It's standard practice and makes logical sense to send out the good stuff for review. Thus we have to be careful to temper expectations a little I think. We won't know truly know the limits till its in the hands of consumers.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |