Nvidia Pascal GPU has 17 billion transistors

you are right the 980 ti is faster but is it 600.00 more faster worth it over my 980s
NOPE

"is it 600.00 more faster worth it"

lol. are you an alien? you really seem like you'd be an alien.
 
I am also assuming that the interposer/HMB transistors are included in that 16 billion transistor count which would make it somewhat less impressive.
 
where are you getting that 600 dollar figure from

After I am lucky enough to get 400 ea for my 980s and rebuying 980ti at 679 ea ..not worth it
Forgot to mention subtract 15% for sale costs. Yea so no thanks . I bought and returned two 980ti in sli over clocked them. Was not impressed over my 980s overclocked. Will wait for pascal and more matured skylake and mobos,ddr4 speeds to reach 4000+ soon
 
Volta is what you want. Pascal is the middle point between the current gen and Volta. I predict similar results to what AMD Fury X is compared to the 290x in terms of performance (980ti vs Pascal high end)
 
Volta is what you want. Pascal is the middle point between the current gen and Volta. I predict similar results to what AMD Fury X is compared to the 290x in terms of performance (980ti vs Pascal high end)
And I think you are laughably mistaken.
 
Fun game rofl... Thats why video card manufactures and reviewers use it as a baseline to evaluate the performance of video cards. K|ngp|n for example ring a bell? On to your other comment... My ti was not a poor overclocker and was average to what the 980 ti can do on core , 1506-1533 is respectable ... Do some research.. On to your other comment.. Across the board evga, gigabyte, Msi, non reference have been hot , loud, and not stable like people want it, again do some research... At 4k yah you will have an improvment but not much at 2k where i want , in fact for almost 700 bucks a single card should be giving me 20-30fps more but its not. 10-15 at most is more like it. Not worth it in my book , i will put my money towards skylake and 8 physical cores and get pascal . If people have 980s in sli and have a 4k monitor yah i would upgrade. Still waiting for 4k IPS at 144hz gsync though, i will give it 6 month to a year. No bad expereience here i did several runs in SLI with a highly overclocked TI and was sad to say the least for the money nvidia wants for these cards. Like i said earlier, hot idle, hot load, hot case , everything was hot but the fps . Its obvious the non reference heatsink sucks and is barely suitble for overclocking

You know you are at [H]ard right?

Try asking Brent or Kyle why they do NOT use 3Dmark...the rest of you ramblings are just as funny...in a sad way.
 
you are right the 980 ti is faster but is it 600.00 more faster worth it over my 980s
NOPE

You sound like you want to make yourself feel better for buying an out of date 980.

Lol.

There is another guy on these forums ---you remind me of him--- who was insisting to others that Asus's forthcoming 144hz IPS g-sync display was a sure thing for an August release , no worries, but then went into denial when confronted with information to the contrary, I believe it was a Q4 release that was stated? If I recall that guy went on to buy the Acer not long after that discourse and then turned around and began chastising others waiting on the Asus, claiming it wont be out till the end of the year at best. You remind me of that guy. Gosh, what was his name?
 
*Facepalm @ thread*

back on topic, I hope SLI 970's will hold me out until pascal, probably won't make the same mistake with the 970's, and wait for the full Pascal chip
 
You sound like you want to make yourself feel better for buying an out of date 980.

Lol.

There is another guy on these forums ---you remind me of him--- who was insisting to others that Asus's forthcoming 144hz IPS g-sync display was a sure thing for an August release , no worries, but then went into denial when confronted with information to the contrary, I believe it was a Q4 release that was stated? If I recall that guy went on to buy the Acer not long after that discourse and then turned around and began chastising others waiting on the Asus, claiming it wont be out till the end of the year at best. You remind me of that guy. Gosh, what was his name?

Wow you make a great troll. I commend you for your efforts. Now forgive me while I enjoy my sig until pascal is released which will ultimately blow the doors off your literly hot GPU . BTW the Acer you talk about gave me a better wow factor than 980s ti in sli. Ooo it felt so good to see almost 1400.00 back on my credit card when I got rid of those hot steaming loads of crap .
 
Last edited:
*Facepalm @ thread*

back on topic, I hope SLI 970's will hold me out until pascal, probably won't make the same mistake with the 970's, and wait for the full Pascal chip


Agreed
You will see a nice bump. You could consider selling both your 970s and just go with one 980 ti.. If I were you wait for MSI lightning ti version it will blow the doors off the current MSI cards ,especially once the modded bios is released for it. However by the time this happens pascal will be out within 3-4 months and AMD will release an updated Fury. Nvidia should have just released a titan 6gb for 679.00 not no cripple
 
Last edited:
You know you are at [H]ard right?

Try asking Brent or Kyle why they do NOT use 3Dmark...the rest of you ramblings are just as funny...in a sad way.

OK so I guess millions and millions of gamers who use 3d mark are all wrong. I forgot guru3d is completely wrong so advoid that site all together for video card reviews. They use 3dmark!!
no your Funny in a very sad way. Are you like a lemming that would Jump off a bridge if everyone else did?
 
Last edited:
With all due respect, canned benchmarks have never represented the user experience as it does not take into account the variables which present themselves at random instances in each game.

Believe what you want, that is your prerogative. Yet, user experience pertaining to their system config reins supreme over anything else.

OK so I guess millions and millions of gamers who use 3d mark are all wrong. I forgot guru3d is completely wrong so advoid that site all together for video card reviews. They use 3dmark!!
no your Funny in a very sad way. Are you like a lemming that would Jump off a bridge if everyone else did?
 
With all due respect, canned benchmarks have never represented the user experience as it does not take into account the variables which present themselves at random instances in each game.

Believe what you want, that is your prerogative. Yet, user experience pertaining to their system config reins supreme over anything else.


I understand what you are saying , however i do not run a 3-5 year old processor with my main OS on a 5400rpm hard drive. Not to mention the fact that 3dmark also catorgorizes your system score to millions of other users with your system specs.

Also when you study enough benchmarks from review sites that right there gives you a good baseline how your system should perform from the results of 3dmark default and extreme

Having both of this information as you call it, " canned benchmarks" will help a person determine how a video card performs overall.
 
OK so I guess millions and millions of gamers who use 3d mark are all wrong. I forgot guru3d is completely wrong so advoid that site all together for video card reviews. They use 3dmark!!
no your Funny in a very sad way. Are you like a lemming that would Jump off a bridge if everyone else did?

There are millions and millions of clueless gamers...bad argument.
Again...why do you think [H]ard (amongst other sites) never does any 3Dmark?

Couldn't it be that the number are useless in comparing real world gaming performance between SKU's...now could it?

Try googling HD 2900 XT 3Dmark...and discover for yourself why 3Dmark is useless...

EDIT
Like this:
http://www.guru3d.com/articles-pages/jetway-radeon-hd-2900-xt-crossfire-review,13.html

In 3Dmark05 we push a score of 17936 points with a HD2900 XT; which is a new record for a single card. We see a lovely 10768 points for 3DMark06. Crossfire is pushing 19180 points in 05 and 13796 on 3Dmark06.

It really is a nice score, but makes one wonder how reliable the score actually is.

Gaming was a different experience...WAY different...the 3Dmark scores..utter useless!

Nothing has changed to alter that...besides, have you completed 3Dmark yet...how was the game? The replay value?
 
Last edited:
There are millions and millions of clueless gamers...bad argument.
Again...why do you think [H]ard (amongst other sites) never does any 3Dmark?

I guess K|ngP|n is clueless and so are all the video card manufacturers like Asus, MSI, Gigabyte,Evga who advertise 3dmark as a baseline to how their video cards perform.

your wrong


Couldn't it be that the number are useless in comparing real world gaming performance between SKU's...now could it?

i have no idea what you are trying to say here. If you can elaborate because I am having a hard time. Its not hard when 98% of the cards run at reference speed and overclock the same and so do the non reference cards. Unless you are talking about the MSI lightning and using exotic cooling . Your comment is meaningless to 98% of the overclocking community.


Try googling HD 2900 XT 3Dmark...and discover for yourself why 3Dmark is useless...

EDIT
Like this:
http://www.guru3d.com/articles-pages/jetway-radeon-hd-2900-xt-crossfire-review,13.html


Im good , i already know about the AMD fiasco with 3dmark and i am talking about current gen cards. Not cards made 6 years ago.



Gaming was a different experience...WAY different...the 3Dmark scores..utter useless!

Nothing has changed to alter that...besides, have you completed 3Dmark yet...how was the game? The replay value?


again see comment above.
And yes i have completed both overclocked benchmarks using SLI in 980 and 980 ti

To me it was not a worth wild investment just because of the power usage, percentage gain at 2560x1440 on a single card and SLI. Games that do struggle on a single card can be circumvented with SLI and all the major games about 2-3% use SLI where my 980 overclocked cant handle 60fps.

so give it up already i am just not impressed with the 980 ti performance and characteristics of the hot idle/load temps vs my 980s in SLI.
Pascal will improve on all these in less than 6 months and put the 980 ti to its grave.

If I were running 4k or upgrading from a 970 SLI yeah a single ti would be a good choice,
 
Last edited:
I guess K|ngP|n is clueless and so are all the video card manufacturers like Asus, MSI, Gigabyte,Evga who advertise 3dmark as a baseline to how their video cards perform.

your wrong




i have no idea what you are trying to say here. If you can elaborate because I am having a hard time. Its not hard when 98% of the cards run at reference speed and overclock the same and so do the non reference cards. Unless you are talking about the MSI lightning and using exotic cooling . Your comment is meaningless to 98% of the overclocking community.





Im good , i already know about the AMD fiasco with 3dmark and i am talking about current gen cards. Not cards made 6 years ago.






again see comment above.
And yes i have completed both overclocked benchmarks using SLI in 980 and 980 ti

To me it was not a worth wild investment just because of the power usage, percentage gain at 2560x1440 on a single card and SLI. Games that do struggle on a single card can be circumvented with SLI and all the major games about 2-3% use SLI where my 980 overclocked cant handle 60fps.

so give it up already i am just not impressed with the 980 ti performance and characteristics of the hot idle/load temps vs my 980s in SLI.
Pascal will improve on all these in less than 6 months and put the 980 ti to its grave.

If I were running 4k or upgrading from a 970 SLI yeah a single ti would be a good choice,

Again, there is a lot of pagehits from millions of ignorant gamers...dosn't alter the fact that 3Ddmark is useless in comparing different SKU's.

You could use it as a primitive gauge for assessing OC improvements on a single system.

But besides that...useless.
 
Again, there is a lot of pagehits from millions of ignorant gamers...dosn't alter the fact that 3Ddmark is useless in comparing different SKU's.

You could use it as a primitive gauge for assessing OC improvements on a single system.

But besides that...useless.

Ignorance is bliss and it is written all over your face
 
Do we even know anything about Volta?

147449-004-CFA07042.jpg


Volta....
 
Alessandro Volta held the chair of experimental physics at the University of Pavia for nearly 40 years and was widely idolised by his students.[1]

Volta was credited with the invention of the first usable battery

The battery made by Volta is credited as the first electrochemical cell. It consists of two electrodes: one made of zinc, the other of copper. The electrolyte is either sulfuric acid mixed with water or a form of saltwater brine. The electrolyte exists in the form 2H+ and SO42−. The zinc, which is higher in the electrochemical series than both copper and hydrogen, reacts with the negatively charged sulfate (SO42−). The positively charged hydrogen ions (protons) capture electrons from the copper, forming bubbles of hydrogen gas, H2. This makes the zinc rod the negative electrode and the copper rod the positive electrode.
600px-Voltaic_pile.svg.png


Now lets take a look at the next Nvidia card Volta

NVIDIA-Volta-GPU-635x357.png



upcoming GPU technology will offer roughly 10 times the performance of Maxwell-powered GPUs in terms processing power related to "deep learning" apps.
GTC - Nvidia Pascal GPU roadmap 2015
According to Huang, Pascal will also bring three key technologies not before seen in previous GPU generations to the market in 2016, including 3D-stacked RAM, which will result in three times memory bandwidth when compared to Maxwell, and NVLink, which will enable the connection of multiple GPUs with very high speeds.
Huang claimed NVLink will outperform PCI-E by up to five times, and be able to run up to eight discrete GPUs in a single machine, rather than today's limit of four.
Better still, Huang said the GPU will be 10 times faster than today's Maxwell-based GPUs, running up to 32GB of RAM - almost three times that seen on 12GB Maxwell-based cards - although he added that this was based on "very rough estimates".
The third and final technology introduced with Pascal will be "mixed precision" capabilities, which Huang said were a first for the company.


It obviously has a resemblance for the Volta Battery where materials are stacked or some sort of engineer design that has the ability to perform a certain way .


As you can see the Volta with Stacked DRAM.
Volta.jpg


http://cdn.wccftech.com/wp-content/uploads/2013/03/NVIDIA-Volta-GPU.png

http://images.anandtech.com/doci/8727/Processors.jpg
 
Isn't the stacked RAM just the HBM that's going to be on Pascal?

If so, it seems we know nothing about Volta. Other than the dude himself.
 
The previous roadmap had Volta coming after Maxwell, and it brings stacked RAM. But in the new roadmap, Pascal will be the next architecture bringing stacked RAM. So nothing has been revealed yet regarding Volta from the new roadmap.
 
Ad hominem...I guess you ran out of real arguments.
Nice own goal :)

nope you dont make any sense at all

If that was the case 3dmark would have gone out of business and no one would ever use it online, in reviews to base the performance of a video card


you still running a 6 year old AMD card called the XT 2600 ?
 
Another vote for 3DMark being pretty but completely useless in gauging real-world game performance. Bad grammar and fragmented thought processes don't change that. :D
 
Last edited:
You're certainly entitled to your opinion and to care about your 3DMark Score... but it's meaningless to me. You linked a page that doesn't tell me anything.

It's like going to a doctor and being told your eGFR is 23 with 1.1g/24H proteinuria despite maximal medical therapy, and on top of that your PTH is 200 and VD3 is 19.

Doesn't really mean anything to the actual person. Much like a 3DMark score.

But if you tell that person they're a year away from needing dialysis and you need to put a fistula in, all of the sudden that's actually useful information.

Much like a real-world game benchmark.

Don't really care how many 3DMark points a card gets.. I mean they're shiny and pretty and sound nice. But it doesn't mean anything to me. Doesn't tell me how many FPS I'm going to get in <insert game here>.

But again, you're entitled to your opinion. :D
 
Show me some concrete evidence that 3dmark is not a good tool to evaluate overall system performance,primarily the video card. Stop giving nonsense medical anologies . Can you explain technically and provide a reliable source on the internet why 3dmark is a useless benchmark in overall system performance? BTW we are not on a medical forum. I don't think you understand how 3dmark works
 
I guess EVGA doesnt know crap about benchmarks then

http://hexus.net/tech/news/graphics/85100-trio-3dmark-world-records-busted-evga-kingpin/

:rolleyes:
You should startup your own video card company so you can convince everyone that 3dmark is useless:D

Like I said....there are milllions of uninformed gamers...you seem to fit the description.

If 3Dmark had any real world value...why do you think sites benchmark games..and not just put out a 3Dmark nuber and call it a day?

Here is a little history for you about 3Dmark:
http://hardforum.com/showthread.php?t=816850

3Dmark has zero corrolation with real world gaming performance, like it or not.

TL;DR
They use 3Dmark because of uninformed posters like your self.
 
Show me some concrete evidence that 3dmark is not a good tool to evaluate overall system performance,primarily the video card. Stop giving nonsense medical anologies . Can you explain technically and provide a reliable source on the internet why 3dmark is a useless benchmark in overall system performance? BTW we are not on a medical forum. I don't think you understand how 3dmark works

Just because you don't understand something doesn't make it nonsensical. :eek:

3DMark is useless to me. Telling me that a Kingpin card gets 25,000 3DMark points tells me exactly that: a Kingpin card gets 25,000 3DMark points. It means nothing. Show me a comparison between the Kingpin card and another 980Ti running Witcher3 and I can actually use that information to decide whether or not the Kingpin card is worth buying.

Very simple. :)

But again, you are entitled to your opinion. Perhaps we are all just too simpleton to understand the braintrust that you bring to the discussion. Snicker.
 
3DMark is good for comparing different GPUs IN THAT SPECIFIC BENCHMARK. That's why reviews have multiple benchmarks so people can see multiple points and how they perform to form the overall picture. Same thing with CPUs. Just running one benchmark or holding its results as the best gauge of a system's performance is not the way to go.

Besides it's well known that different benchmarks are more or less taxing to the GPU. You can easily run 3DMark, Heaven or Valley with much higher clocks than what is stable in Witcher 3 or GTA V. While I know benchmarks are easy to run and give you neat numbers at the end, they don't reflect the games you actually play using your card.
 
3DMark is good for comparing different GPUs IN THAT SPECIFIC BENCHMARK. That's why reviews have multiple benchmarks so people can see multiple points and how they perform to form the overall picture. Same thing with CPUs. Just running one benchmark or holding its results as the best gauge of a system's performance is not the way to go.

Besides it's well known that different benchmarks are more or less taxing to the GPU. You can easily run 3DMark, Heaven or Valley with much higher clocks than what is stable in Witcher 3 or GTA V. While I know benchmarks are easy to run and give you neat numbers at the end, they don't reflect the games you actually play using your card.

I would agree

but some people on here say 3dmark is completely useless as a tool to benchmark. If anyone knows 3dmark has several tests to include physics which will max out your cards potential clock speeds. It is also a known fact the higher your 3d mark score whether it be default or extreme will net you an overall higher score than the card that has a lower gpu score. I would agree and have mentioned this before looking at reviews so see those multiple benchmarks in game is a good tool to reference how the video card performs. But to say 3dmark is useless benchmark is just dumb .
 
Back
Top