maxwell Titan 34% faster

[Kaapstad]

It's not a worthy upgrade from a 780Ti, unless you're desperate for the extra VRAM in SLI. I've yet to see any test where a 780Ti is quicker though.

You have not looked very hard then lol

Remember these mid range 9 series cards have some serious weaknesses @4K.

 

[Silent_Scone]

Nobody is talking about 4K apart from you Kaap, look past your nose. You'll see the screen once in awhile.

Here's a memory intensive bench for you. Remember this? One that took copious amounts of AA at any resolution for the 290x to surpass 780Ti.

Very weak here clearly. Maxwell drivers still need work, this is something you fail to understand at every turn. Much like with X99, but that's crap too isn't it Kaap.

http://36646d87786feafc0611-0338bbb...dn.com/images/HNH2nuGD-nFS.878x0.Z-Z96KYq.jpg

No hotlinked images, if you want to post an image please rehost on your own webspace. Thank you.

This card is a successor for people who need to upgrade. If anyone needs to look harder Kaap, it's you for buying them when you have a 4K monitor.

In fact, if I recall you were told by none other than myself and another that it's not worth buying them for 4K, so I can only but laugh mate, sorry.

 

[Kaapstad]

It's not a worthy upgrade from a 780Ti, unless you're desperate for the extra VRAM in SLI. I've yet to see any test where a 780Ti is quicker though.

Nobody is talking about 4K apart from you Kaap, look past your nose.

Well actually you opened it up to 4K with your previous post, sir you are busted.

As to 4K performance with the 980s I really don't care as this is not the reason I bought them, I already knew it was going to be poor even before I got them out of their boxes. This does not take much working out, lets see why

4gb VRAM - poor for 4K
256bit bus - poor for 4K
No Vapor chamber cooler - poor for 4K

Having said that NVidia never intended for these mid range cards to get heavy use @4K so maybe I am being a bit harsh pointing out their weaknesses.

For the purpose I bought my GTX 980s I am very pleased with them.

Now you bring up X99 I have a question for you - when are you going to get your DDR4 tweaked and tuned properly as at the moment the only reason your physics scores look good is because they are inflated by a high CPU clockspeed (which is down to the silicon lottery), I am only mentioning this because you are very good at running other people down but you don't seem to do so well on the platform yourself.

One last bit of advice, try not to pick fights with every second person on the forums as this will not do you much good in the long run.

 

[Silent_Scone]

I've not once mentioned 4k you baffoon. 3Gb VRAM is a bottleneck at far lower than 4k in some instances.

As for the rest, utter rubbish as per. Not sure what you mean by tweaked and tuned properly. The only reason my physics score is better is because unlike you, I do research, change and chop where needed, but most importantly, I'm not a cranky old man who shouldn't be buying new platforms if you're not willing to put the time into it.

You forget how long x79 took to mature, and your dig at my memory 'tuning' only further proves your ignorance on half the things you talk about, mate.

I'll humour you as usual though.

I don't tune the memory during 3d runs for several reasons. Most importantly, it makes little difference on this platform. Secondly, I'm not comfortable using the voltage required to do so.

Maybe I will run it down when you can beat me


Why don't you ask Vince the same question as to why he's running such a low uncore and Cache latency whilst at 2T. I'm sure you could teach him a thing or two as well.

http://www.kitguru.net/wp-content/uploads/2014/11/evga_kingpin_geforce_gtx_firestrike_-_single.jpg

If the image is larger than 1280 pixels wide please resize, link only or put in spoiler tags. Thank you.

 

[IT Troll]

Disappointing to hear that the "Ti" is going to be a cutdown Titan. That was what was so great about the 780Ti, full fat chip where it counted.

 

[Orangey]

GM200 da best
GM204 *hock ptooie*

 

[Disco_P]

That's a very simple way of looking at it. It's not just about power bills. What do you think it's like having three or four Titans in a loop in Summer? A full fat GK110 can consume up to 400w on a reference board each.



That energy isn't just going into your bills it's dispersing into your room. 980s consume even once overclocked little more than 200w. So they might still not be 'worth' it, but those same people probably aren't power users. H100 cooler and a single 290p type deal.

It does all this whilst being on the same process, and faster than Kepler, but it's not high end just because Dave and Dell Boy know there's something else planned. If you took that outlook every time a company that uses fabrications said anything what so ever, you'd spend your whole life waiting...like some chump

The most laughable part of if all though is guaranteed, a large majority of these same people won't be willing to pay the prices for what Nvidia have shelved. So when push comes to shove I'm not sure who's the bigger idiot, people who've bought GM204, or those so desperately clinging on to the unknown. I'll get what ever comes just because I don't mind, it's something I do because I enjoy it. I don't try to tell Nvidia what their products are because I think I know better than they do, regardless of what Rajif from India leaked from a shipping manifest.

As I said, I own a single 780 - not 3 or 4 titans.

And I don't need to look at shipping manifests to know that nvidia and amd are going to release faster cards.

 

[Silent_Scone]

You also said something about looking at your power bill though.

Can't tell emphasis through text, maybe you were being funny.

 

[bru]

As to 4K performance with the 980s I really don't care as this is not the reason I bought them, I already knew it was going to be poor even before I got them out of their boxes. This does not take much working out, lets see why

4gb VRAM - poor for 4K
256bit bus - poor for 4K
No Vapor chamber cooler - poor for 4K

ER.....
Ok I can understand the 4GB Ram and the 256bit bus, but surely the cooler regardless of what type it is makes no difference to the GPU for 4K.
Or are you saying that they throttle without a vapour chamber cooler. If that is the case turn the fans up, add additional ventilation. You could always water cool them, you probably have haven't you ?

 

[Kaapstad]

ER.....
Ok I can understand the 4GB Ram and the 256bit bus, but surely the cooler regardless of what type it is makes no difference to the GPU for 4K.
Or are you saying that they throttle without a vapour chamber cooler. If that is the case turn the fans up, add additional ventilation. You could always water cool them, you probably have haven't you ?

4 up with the fans on maximum they won't even manage a lap of Heaven 4 without crashing if you overclock them and the number of artifacts goes off the scale.

As soon as I put waterblocks on what a difference, no artifacts and very stable. Where the 4K reference comes in is that the cards produce more heat at the resolution as they are under max load more of the time.

I don't want people to get the impression that I don't like the 980s as I have used them a lot more than any of my other cards recently. What I do have an issue with is people who bury their heads in the sand and pretend they are perfect because they own 3 of the cards.

 

[Kaapstad]

I've not once mentioned 4k you baffoon. 3Gb VRAM is a bottleneck at far lower than 4k in some instances.

If you want to make a good argument you should not let it become personal lol.

As for the rest, utter rubbish as per. Not sure what you mean by tweaked and tuned properly. The only reason my physics score is better is because unlike you, I do research, change and chop where needed, but most importantly, I'm not a cranky old man who shouldn't be buying new platforms if you're not willing to put the time into it.

Is it a crime to be old and using a PC lol.

I actually put a lot of time into it, this is why I know you are getting poor performance on your memory for example.

You forget how long x79 took to mature, and your dig at my memory 'tuning' only further proves your ignorance on half the things you talk about, mate.

I'll humour you as usual though.

I don't tune the memory during 3d runs for several reasons. Most importantly, it makes little difference on this platform. Secondly, I'm not comfortable using the voltage required to do so.

I know exactly how long it took for X79 thanks and it was a lot better than X99

As to tuning DDR4 on X99 making little difference for futuremark benches, this is one of the reasons why the platform is rubbish as on other platforms it does make a difference. I also notice memory performance on X99 hexcores is better than the 5960X taking into account the number of cores, have intel messed up here.

Maybe I will run it down when you can beat me

As I have lost count of the number of times I have beaten your scores with DDR3 often with me using a slower CPU as well I don't think I will take this seriously.

Why don't you ask Vince the same question as to why he's running such a low uncore and Cache latency whilst at 2T. I'm sure you could teach him a thing or two as well.

Why don't you ask him for a few tips, he knows a lot about chilling.

 

[Dave2150]

Now you bring up X99 I have a question for you - when are you going to get your DDR4 tweaked and tuned properly as at the moment the only reason your physics scores look good is because they are inflated by a high CPU clockspeed (which is down to the silicon lottery), I am only mentioning this because you are very good at running other people down but you don't seem to do so well on the platform yourself.
)

Well, he seems to be doing way better on x99 than you are, as you admit in your post. His scores beat yours, you make yourself look rather silly by saying he doesn't "do so well" when he's doing better than you.

 

[Kaapstad]

Well, he seems to be doing way better on x99 than you are, as you admit in your post. His scores beat yours, you make yourself look rather silly by saying he doesn't "do so well" when he's doing better than you.

Silicon lottery.

 

[Silent_Scone]

So, because Futuremark, a predominantly GPU bench, doesn't scale well when running lower latency...this makes the platform crap? You say silicon lottery but I notice you're running 4Ghz in a lot of the tests you run, and 4.5 for benching.

You cannot tell me that your chip isn't capable of running at the very least 4.2ghz for daily use? I don't think it's so much of the silicon lottery as it is you.

I don't recall you beating my scores at all with DDR3, so I'm not sure what you're talking about there. I can dig some up though if you really want to go there.

It's really difficult to reply to you without being condescending when you just say the silliest things, sorry. I almost feel bad now - it's the weekend, lets just let it be shall we.

 

[Gerard]

 

[Disco_P]

This thread has gone full potato.

And some people might want to look up strawman fallacy between grabbing benchmarks.

 

[Silent_Scone]

This thread has gone full potato.

And some people might want to look up strawman fallacy between grabbing benchmarks.

The whole subsection is a dumping ground with absolute nonsenses posted, the full potato comes when you pull people up for talking nonsense in the first place.

Much prefer other boards where people talk sense.

 

[TwsT]

The whole subsection is a dumping ground with absolute nonsenses posted, the full potato comes when you pull people up for talking nonsense in the first place.

Much prefer other boards where people talk sense.

don't let the door hit you where the good lord split you,


You've pretty much argued with yourself the entire thread.

 

[Silent_Scone]

don't let the door hit you where the good lord split you,


You've pretty much argued with yourself the entire thread.

Well you probably know what they say...about beating with experience...

I'm not the only one who sees through Kaaps jibber
jabber, but when he tries calling me out with a blanket statement that makes no sense, no matter how fluffy he is I get a bit tired of it.

Your door needs oiling btw

 

[TwsT]

that's fine but none of anything you said was relevant at all to your initial comment about nvidia needing to fix drivers. Which was probably the only thing you said that had any chance of being on topic.