technical capabilities of the architecture? are not that good for the size & resources that it uses for what it is, its should be doing them much better.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
ATI or NVIDIA? Which is Better?
- Thread starter Xientie
- Start date
technical capabilities of the architecture? are not that good for the size & resources that it uses for what it is, its should be doing them much better.
As I said it would probably be the case but theres several good reasons why AMD didn't take that approach, for all we know there could be serious hinderances to performance.
How does that have any relevance?
again what relevance does this have to which arcitecture is the more advanced in its technical capabilities?
Soldato
- Joined
- 8 Aug 2010
- Posts
- 6,453
- Location
- Oxfordshire
It should really be compared to the 5870.
AMD doubled up the Via's to add redundancy which took up die space, where as Nvidia strategy was to disable a faulty cluster, the disabled cluster is effectively Nvidia's form of redundancy.
A more transparent comparison of the architectures would be to compare Juniper Vs GF106 as Nvidia didn't have to disable any 'clusters'.
http://www.anandtech.com/bench/Product/180?vs=162
Relevance is how well something is done = performance.
You keep going on about the architecture like its the most important factor regardless of the performance of that architecture.
The performance of an architecture is the most important part in relation to others architectures no matter how old or new they are.
So saying that's its new like its a plus = absolutely nothing.
Femi=Betamax ATI= VHS.
Betamax having the slightly better picture performance did not save it because it used twice as much resources = tape as VHS.
Last edited:
Your completely missing the point - I'm not saying either architecture is good or bad.
Soldato
- Joined
- 8 Aug 2010
- Posts
- 6,453
- Location
- Oxfordshire
You keep spouting Fermi is a generation ahead, or a step forward in the 'GPU trend' when it's only true with regards to GPGPU (which we previously agreed was irrelevant in this thread), and in fact a step backward with regards to gaming.
How is it a step backwards with gaming?
Soldato
- Joined
- 8 Aug 2010
- Posts
- 6,453
- Location
- Oxfordshire
People don't care about the architectural changes if its not backed up by its performance.
Soldato
- Joined
- 8 Aug 2010
- Posts
- 6,453
- Location
- Oxfordshire
Because it's inefficient and costly, and can't compete with rival cores in it's size or weight class.
I'm saying that in terms of technical progress Fermi is the more advanced design, a few pages back someone claimed it was behind Evergreen.
Its like comparing a propeller plane to a jet - both will generally get you to your destination in a similiar amount of time but no ones going to say they are the same generation design.
Its like comparing a propeller plane to a jet - both will generally get you to your destination in a similiar amount of time but no ones going to say they are the same generation design.
But its outperforms its equivalents in the same market segment, is currently decently cost effective to the consumer for performance and has a full set of features. So what difference does it make?
Soldato
- Joined
- 8 Aug 2010
- Posts
- 6,453
- Location
- Oxfordshire
That's true, but with regards to gaming Fermi is the propeller plane.
More advanced architecture but not as good because it needs to be twice as big so what does it matter that its more advanced in theory.
Its no better than the 4+1 cluster of ATi being more advanced in theory but in reality it was hard to use all of it so theatrical performance gain by having the 4+1 cluster became irrelevant.
Last edited:
Burning up on re-entry.
Soldato
- Joined
- 8 Aug 2010
- Posts
- 6,453
- Location
- Oxfordshire
It doesn't make a difference to the consumer right now per se , but Nvidia's current architectural strategy is unsustainable, and in order to keep it's products competitive Nvidia will have to lose money if it doesn't want to lose more market share.
If that happens there will be less money spent on R&D and the graphics cards business will be less competitive as time goes on and so will spiral into a viscous cycle, consumers will be hit with higher and higher prices due to lack of competition until the point Nvidia might one day cease to exist (at least in it's current form), and that would definitely affect consumers and not in a good way.
It would be naive to think that Nv will be able to fall back on the professional or HPC market, as the only reason it is currently able to prop up Nvidia is because there is no competition and Nvidia can charge what they like, and the huge development cost's are offset by Geforce.
What do you think is going to happen when Intel with all of it's resources pushes hard into this market like it is beginning to do?
If nothing else it will squeeze margins, and that could be fatal especially if Geforce can't make a profit.
In short, it is short sighted to suggest consumers are not being affected, it's just that the reaction to the action is delayed, but that doesn't mean it won't be any less painful when it is finally felt.
Last edited: