RTX 3080 relative performance data from Digital Foundry's video

[Blademaster91]

5 minutes ago, Exty said:

stop with that meme already it makes zero sense , every gpu before this one pushed the air up in to the CPU already. where do you think your GPU air went? poof?

The previous cards weren't exhausting 300w or more onto the CPU, I think for those on air cooling a partner card would be better, although those look to have cooling slots through the PCB.

[Bombastinator]

9 minutes ago, AluminiumTech said:

Moore's Law Is Dead YT Channel was suggesting that Nvidia wants to dispose of the SUPER naming scheme because customers found it too confusing.

Well I'd say it's bad to buy right now cos we don't know benchmarks and we don't know how RDNA2 will stack up.

3070 coming out October.

 

3060 possibly wayyyy later.

Oh yeah definitely. Nvidia short-changed everyone here.

 

I think the 3080 should have come with 20GB and the 3070 with 16GB leaving the 3060 with 12GB.

 

Unfortunately that's not how the world works and Nvidia will only release a 3080 20GB or a 3070 16GB if AMD competes well enough.

I got the impression the only reason the super thing ever happened was because AMD came out with faster cheaper cards, just like with the ti line.  If AMD comes out with faster and/or cheaper stuff I suspect we will see Ti pop up again. 
The deal seems to be in this market that Nvidia CAN make faster cheaper stuff. They just don’t WANT to. They will only put forth the minimum necessary to bake them a butt hair faster and that’s it.  It’s literally producing the worst product possible.  Not that AMD would be any better necessarily.  They’ve just been doing second fiddle lately. 

Edited September 2, 2020 by Bombastinator

[AlTech]

1 minute ago, Exty said:

you do have to remember that only half of the air from the gpu is comming inside the case, it works like a blower and a conventional fan

No not really. Each of the FE cards has 2 fans. The 3090 and 3080 have one on top and one on bottom but they take in 100% of the air from the case and spit out the hot air back into the case to be disposed of by your case fans.

[AlTech]

Just now, Bombastinator said:

I got the impression the only reason the super thing ever happened was because AMD came out with faster cheaper cards, just like with the ti line.  If AMD comes out with faster and/or cheaper stuff I suspect we will see Ti pop up again. 

Oh yeah definitely, Nvidia never wanted to release SUPER series cards. They did so because of AMD.

[Exty]

Just now, Blademaster91 said:

The previous cards weren't exhausting 300w or more onto the CPU, I think for those on air cooling a partner card would be better, although those look to have cooling slots through the PCB.

most of the heat will be pushed out the case by the blower style side because the pcb is half the size of the card, so i would expect the right side of the gpu only be in charge of the exec heat. hey but what do we know we can only speculate.

[Exty]

17 minutes ago, Blademaster91 said:

The previous cards weren't exhausting 300w or more onto the CPU, I think for those on air cooling a partner card would be better, although those look to have cooling slots through the PCB.

how is this better?

[Gohardgrandpa]

5 hours ago, JediFragger said:

Managed to sell my 2080Ti a week ago *phew* !!

You lucked out big time 

[Juanitology]

44 minutes ago, Shreyas1 said:

I mean in all fairness, 500 bucks will still get you a new card that is better. 

no, that'll happen in october. Some people are willing to pay an extra hundred bucks to get their hands on a card sooner. 

[AldiPrayogi]

5 hours ago, Valentyn said:

I've had it since launch as well. I'm not complaining.

A friend of a friend refused to wait for the event recently, and bought a 2080Ti last week!

F

[AldiPrayogi]

BTW how much do guys reckon a 1660Ti will sell once 3000 series comes out? Looking to sell mine...

[JZStudios]

6 hours ago, Bombastinator said:

Got that digital foundry video pushed on me earlier.  It bugged me.

So you saw 1 video years ago and made a bad judgement, thus this guy is a shill and not trustworthy.

Maybe look at any video they've done in the past 5 years. Yeah this video wasn't great. Why we needed him to tell us he had the card plugged into the TV I don't know. Doesn't mean it's invalid, nor does it mean he's a shill.

[Bombastinator]

11 minutes ago, JZStudios said:

So you saw 1 video years ago and made a bad judgement, thus this guy is a shill and not trustworthy.

Maybe look at any video they've done in the past 5 years. Yeah this video wasn't great. Why we needed him to tell us he had the card plugged into the TV I don't know. Doesn't mean it's invalid, nor does it mean he's a shill.

The Mac pro was “years ago”? I think it was a good bit more recent than that.  Also the last time I saw a video from this guy.  It is reasonable that he produces more of them.  I don’t see them though.  I’ve merely had two of them shown to me. 

 

I never said he was.  I said it seemed that way because of what I have seen. 

 

 From what I have seen, it doesn’t look like it wasn’t either.  You think I should watch five years of video before I dare to state what I saw?  I see what gets reposted.  What I’ve seen reposted are those two things.  Do people only repost videos of his that are unrepresentative of what he does?   Or do people only repost stuff of his when he is the first person to get equipment?  
 

Both times he said “I do not do this thing” and then did that thing.  Perhaps most of the time he doesn’t.   I don’t know.  I can tell you I did not see it as confidence building given what I have seen. 
 

Clearly you like the stuff he produces.  Good for you.  If you’d like to make an actual case as to why he should be listened to I’m ready to hear it.  You haven’t done that though.  You’ve merely  accused me indirectly of saying stuff I didn’t. 

[Guest]

2 hours ago, JZStudios said:

Why we needed him to tell us he had the card plugged into the TV I don't know.

He talked a bit about it in the next video. He basically received the card from nvidia without any technical informations (cuda cores, wattage, architecture) so he didn't really believe those numbers himself. It wasnt until the september reveal that he himself understood why the card he got to test was this fast.

[RejZoR]

5 hours ago, Exty said:

how is this better?

Because that's not how hot air rises from 3 fan cards. It rises from the side, not directly from the card. And usually there are 3 fans in the front, this 2 fan intake setup is just bizarre with empty space where third fan should be... Which mixes the air before it goes at CPU. Which would also make NVIDIA's new cooler less of a problem to begin with. Not to mention top exhausts which basically all cases have these days, even cheapo ones.

[straight_stewie]

10 hours ago, Mark Kaine said:

It's also typical for NV to milk their customers as much as possible,  which also partly explains the relative "tame" prices for the 3xxx series, so far. 

That's only been the case for the 10xx and 20xx series.

This "new" pricing is a return to their previous normal. Perhaps they've learned something?

[leadeater]

8 hours ago, Exty said:

how is this better?

Well if that ever existed then it wouldn't be, but no card has ever done that. AIB cards of similar design to that the fans intake in the front of the card (facing down) and push air out the sides in to the case, which mixes and equalizes so you are raising the case ambient but you are not blowing hot air directly in to the CPU heatsink.

 

This is no different to an electric fan heater and you getting hotter sitting in front of it versus sitting away from it, same energy, just not directly applied to you.

 

It's really not going to be that big of a problem, for AMD CPUs only slightly as an extra 5C is 25MHz less, if you could event notice that.

[Bombastinator]

14 minutes ago, leadeater said:

Well if that ever existed then it wouldn't be, but no card has ever done that. AIB cards of similar design to that the fans intake in the front of the card (facing down) and push air out the sides in to the case, which mixes and equalizes so you are raising the case ambient but you are not blowing hot air directly in to the CPU heatsink.

 

This is no different to an electric fan heater and you getting hotter sitting in front of it versus sitting away from it, same energy, just not directly applied to you.

 

It's really not going to be that big of a problem, for AMD CPUs only slightly as an extra 5C is 25MHz less, if you could event notice that.

I do think this seems likely.  The thing that makes me think so is modern CPUs output a lot less raw BTUs  than the old ones, but many cooler designs are 20 years old.  They’re designed to deal with a lot more heat output than they get.  The problem with new CPUs isn’t exhausting waste heat it’s collecting it to exhaust in the first place.  The air has to travel up for a good distance which means that it will mix with case air.  
 

Time and testing will tell of course. 

[Valentyn]

10 hours ago, porina said:

The Titan V does have a party trick that makes it extremely interesting in some compute use cases: decent FP64 performance. Consumer cards have had this crippled for a very long time. The Radeon VII was the only exception, and I don't really consider it a real consumer card anyway since it was basically a Pro card rebadged. Titan V is ball park double the performance of a Radeon VII. The VII silicon could near enough match the Titan V if AMD didn't intentionally cripple it to preserve their pro offerings.

Yup! Chap that bought it said he wanted it for work.

in my mind that means compute or deep learning for one of these.
considering the RTX Titan is heavily capped in FP64 it saves a lot of money if you don’t want a Quadro.

[Valentyn]

28 minutes ago, leadeater said:

Well if that ever existed then it wouldn't be, but no card has ever done that. AIB cards of similar design to that the fans intake in the front of the card (facing down) and push air out the sides in to the case, which mixes and equalizes so you are raising the case ambient but you are not blowing hot air directly in to the CPU heatsink.

 

This is no different to an electric fan heater and you getting hotter sitting in front of it versus sitting away from it, same energy, just not directly applied to you.

 

It's really not going to be that big of a problem, for AMD CPUs only slightly as an extra 5C is 25MHz less, if you could event notice that.

That’s it. I don’t see an issue with my old Noctua really.

 

My ancient 5820k ( 140W TDP stock ) never broke 60C under stress tests, now it might. Oh no!

 

Nevermind until recently I had Titan V, and 2060S in there number crunching, and folding at home without issues. 
 

Only thing that got affected by the extra hot air was the Titan really.

 

Im sure most wont have an issue if they have okay air flow in the chassis. 

[CarlBar]

1 hour ago, Bombastinator said:

I do think this seems likely.  The thing that makes me think so is modern CPUs output a lot less raw BTUs  than the old ones, but many cooler designs are 20 years old.  They’re designed to deal with a lot more heat output than they get.  The problem with new CPUs isn’t exhausting waste heat it’s collecting it to exhaust in the first place.  The air has to travel up for a good distance which means that it will mix with case air.  
 

Time and testing will tell of course. 

You what???!!!

 

Whilst comparing TDP's across that span of time isn't an exact science the top of the line early 2k CPU, (the Athalon XP 3200+), had a TDP of just under 80w. Current mid range AMD CPU's run to 95w's.

 

This is the modern stock cooler, (i.e bargin basement) for AMD CPU's:

 

 

This i believe is the equivalent from the Athlon XP days, (it's been too long for me to remember and google is being a bit weak):

 

 

TDP and Cooler capability has been going up for the last 20 years, not down, and 20 years ago heat[pipe based coolers where all but unheard of.

 

 

 

[KaitouX]

On the cooler:

Sapphire have done the throw hot air towards the top of the case multiple times in the past:

Fury Tri-X

Spoiler

Vega 56

Spoiler

Both which are much the same idea as the cooler on the 3070.

Spoiler

And probably there are others with similar solutions, so it's not like that throwing air at the CPU haven't been done before, what haven't been done is the fan being at the back like on the 3090/3080(as far as I know).

Also one thing is that from the look of the DF video, the FE cooler probably have the fan stop feature, as the GPU fan wasn't spinning even though the system was turned on at the start of the video(1:40 or so). The FE cooler might actually be a good option this time, if it is quiet enough it will probably be a nice option at what seems to be the MSRP for the basic models(?), hopefully the cooler on the 3070 and possibly 3060 also ends up being good.

[leadeater]

45 minutes ago, CarlBar said:

Whilst comparing TDP's across that span of time isn't an exact science the top of the line early 2k CPU, (the Athalon XP 3200+), had a TDP of just under 80w. Current mid range AMD CPU's run to 95w's.

I think that is going back a bit far but from Core 2 Duo/Quad to Sandy Bridge there were massive improvements to power states and idle power usage, and then each revision after Sandy Bridge that also got better. Peak power draw has gone up, and down and all over the place, but generally up however under gaming loads the power draw isn't that great.

 

My 4930k idle power draw is horrible compared to anything modern, ~60W. A current Ryzen mobile U series can outperform what I have without even hitting my idle draw lol.

[Coshy91]

16 hours ago, Valentyn said:

I've had it since launch as well. I'm not complaining.

A friend of a friend refused to wait for the event recently, and bought a 2080Ti last week!

He might be able to hand it back if bought from a store and get his money back.

[spartaman64]

13 hours ago, porina said:

The Titan V does have a party trick that makes it extremely interesting in some compute use cases: decent FP64 performance. Consumer cards have had this crippled for a very long time. The Radeon VII was the only exception, and I don't really consider it a real consumer card anyway since it was basically a Pro card rebadged. Titan V is ball park double the performance of a Radeon VII. The VII silicon could near enough match the Titan V if AMD didn't intentionally cripple it to preserve their pro offerings.

https://www.extremetech.com/extreme/266508-nvidias-titan-v-accused-returning-wrong-answers-simulations

not useful for scientific use apparently unfortunately. unless they fixed it

[Bombastinator]

3 hours ago, CarlBar said:

You what???!!!

 

Whilst comparing TDP's across that span of time isn't an exact science the top of the line early 2k CPU, (the Athalon XP 3200+), had a TDP of just under 80w. Current mid range AMD CPU's run to 95w's.

 

This is the modern stock cooler, (i.e bargin basement) for AMD CPU's:

 

 

This i believe is the equivalent from the Athlon XP days, (it's been too long for me to remember and google is being a bit weak):

 

 

TDP and Cooler capability has been going up for the last 20 years, not down, and 20 years ago heat[pipe based coolers where all but unheard of.

 

 

 

95w.  You are referring to the 3950.  I remember saying once here that a 3950 could be cooled by big air coolers of sizes similar to the D14 and having people jump down my throat saying only water was possible, and how dare I comment on workstation coolers.  They were quite angry.

 

  A 3700x lists as 65w. A3950 lists are 95w a dh14 lists at somewhere between 200-300w

So you’re saying 95w must then be the hottest cpu AMD ever produces then since it’s been according to you a 20 year steady progression?

 

I don’t have AMD numbers handy, just intel.

They’re here:

https://en.m.wikipedia.org/wiki/List_of_CPU_power_dissipation_figures

 

Theres a breakdown on intel and AMD microcomputer CPUs in there I’d you want to dig for it.

Looking at the charts is odd.  AMD has dates but the intel stuff doesn’t.  Intel stuff has wattages but AMD doesn’t. Was this why you chose AMD for your comparison? 

 

 I’ve got a core2 Quad 6600 in my basement with a hyper212 on it. It was rated for something near or over 100w stock. It’s less than 20 years old though. I bought it after I bought my house and I did that in 2000. Hyper212 was newish when I built that machine.  No sure how much less. I want to say 2003 but I’m not sure.  I recall I overclocked the chip.  Don’t remember by how much.  It may not have been 20 years but it’s been a while

 

So perhaps 20 years is not sufficiently exact.  This according to you apparently isn’t about cpu heat progression it’s about the number 20 years old and specificity.  And you base a claim of constant heat progression on that.  You want more specificity?  I can give you granularity. 
 

When did the hyper212 come out? I was thinking something like 20 years.  Perhaps that is too long. Coolermaster had only been going about 8 years at that time.  Maybe it should have been 17 or something. 20 was a round number.   When did single core CPUs first start overclocking to the 5ghz range?  Pentium D era? It was before core2 duo came out.  I remember that because the core 2duos could beat single core celerons running OCed to 5 ghz The core 2duos of the time  were something like 65w iirc.  The big pentiumDs ran much much hotter. Have cpu sizes and temps gone up over the course of micro computing in general? Sure.  There was a time when CPUs didn’t have coolers at all.  Look at a z80. I messed about with those on trash80s and stuff when I was a kid.  CPUs got hotter and hotter and higher ghz and higher ghz until 5ghz was reached where upon they went multicore and the process repeated.  They did two core the same way.  Then four core. All this time nodes were shrinking and transistor counts were growing.  As the nodes went down so did heat but as transistor counts went up so did heat.  It went up for a while but became More of a pulsing after multicore hit.  A pulsing that went in general slightly down.  When the upper range of transistor count was reached but there hadn’t been a die shrink yet, chips get hot.  Intel’s 14nm CPUs are getting pretty hot these days.  They’re rather famously having trouble with the node shrink. Hotter and hotter. They’re well over 100w these days. They’ve made higher wattage chips though.  The interesting bit is they cooled them with smaller coolers and got lower numbers doing it.  SOI really didn’t like going over 75c.  Yet big water radiators weren’t needed.  What’s the difference?  The chips themselves are physically smaller.  Node shrink.  Radiator size has gone up fairly consistently.  Wattage hasn’t though, which was the point I was making.  Old radiators still work fine so they’re still used.  They were engineered for a slightly different situation though. 

 

so yeah.  “What.”

Edited September 3, 2020 by Bombastinator
Tried a typo pass. Found one content error. Post is too long though there May still be typeos