• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Haswell model specs leaked

Page 16 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
IntelUser2000 said:
Search Anandtech for Haswell IDF coverage.

I'm thinking it'll end up at ~1GHz for the quad core GT3. With 640GFlops, that's approximately double what a Ivy Bridge GT2 running at 1.3GHz achieves. Since 20% gain in average will be enough to match Trinity in the mobile, they should be able to do it with GT2 alone. Of course they'll still be behind for Desktop.
Click to expand...
The only problem is AMD will have updated Trinity (richland) which scores 20% more in 3dmark11 versus previous top mobile part 4600m. It's a moving target in Q2 and it will move once more in Q3/Q4 with Kaveri which will widen the gap even more in mobile sector(32nm->28nm, GCN 2.0 vs VLIW5,higher clocks- you get the picture). I still think Haswell will be a massive step in the right direction for intel but I doubt it will pose a threat to Trinity's/richland's iGPU .
 
IntelUser2000 said:
but Haswell increases more than computational performance as well. Like doubling fixed function pipeline performance, and increase in texture performance in some scenarios up to 4x. Also adding the Resource Streamer to reduce driver bottlenecks. Those should help a lot to close up in the weak areas.
Click to expand...

Interesting. EDIT= Makes me wonder that with all those changes Intel has not really promoted the GT's at all. Like, calling them 4th gen IGPU's or something. Just % increases.

inf64 said:
The only problem is AMD will have updated Trinity (richland) which scores 20% more in 3dmark11 versus previous top mobile part 4600m. It's a moving target in Q2 and it will move once more in Q3/Q4 with Kaveri which will widen the gap even more in mobile sector(32nm->28nm, GCN 2.0 vs VLIW5,higher clocks- you get the picture). I still think Haswell will be a massive step in the right direction for intel but I doubt it will pose a threat to Trinity's/richland's iGPU .
Click to expand...

Yeah, it seems that Richland won't be using GCN and still being at 32nm. Haswell will probably get close but Kaveri will show who's architecture is better. Not dismissing that BRW is supposed to be Intel's true next gen iGPU.
 
Last edited:
Kaveri is coming a year later from Richland, not 1-2 quarters. That puts its competition with Broadwell.

At the least Haswell should close up any gap left on the mobile side.
 
True.

But its weird that after the changes that the Haswell iGPU has, it is not being considered as the 4th gen iGPU. You commented on three or four things that make it different than the IVB iGPU's. I find it suspicious of Intel not to promote them, at least, I feel that the IVB iGPU was better promoted.

With those changes, it does seem like a nice competitor. Can't wait to see benchmarks of it. Just hate that the GT3 ain't on an LGA1150...
 
Here is a question.
What happens if Intel manages to bring out a chip which is vastly superior to everything ARM has to offer? Both in terms of performance and power efficiency?
Would companies like Apple and Samsung drop their own CPUs and adapt Intel's?
because if they don't, they will be selling inferior products ( something Apple couldn't live with ) and if they do, they will be at the mercy of Intel for the rest of their lives.
Any thoughts?
 
Well there's still cost, which you didn't mention. Possibly Intel's greatest weakness is with cost, rather than performance, perf/watt or power.

But if they did win on every front, like you suggested, then Intel wins. Period. I don't think there's any room for alternative hypotheses.

But cost is a major and often overlooked concern.
 
GillyBillyDilly said:
Here is a question.
What happens if Intel manages to bring out a chip which is vastly superior to everything ARM has to offer? Both in terms of performance and power efficiency?
Would companies like Apple and Samsung drop their own CPUs and adapt Intel's?
because if they don't, they will be selling inferior products ( something Apple couldn't live with ) and if they do, they will be at the mercy of Intel for the rest of their lives.
Any thoughts?
Click to expand...

They can't. They are still playing catch-up. ARM chips aren't stopping improving, Tegra 4 was announced and The new SnapDragons also.
And then comes markets. ARM is every where. Intel is not.

Personally, the only reason I see ARM as limited in the PC space is OS support. Put Win8/Pro and Mac OS in an ARM, now what?

But sooner or later something will happen, and I like that. Its not boring anymore.

In a few years, maybe in 2014, will see Intel vs Nvidia vs AMD vs Samsung vs Qualcomm vs Apple vs Whatever else. Everyone is doing "cpu" and "gpu" now. We will get better products with all that competition. I think.
 
MightyMalus said:
They can't. They are still playing catch-up. ARM chips aren't stopping improving, Tegra 4 was announced and The new SnapDragons also.
And then comes markets. ARM is every where. Intel is not.

Personally, the only reason I see ARM as limited in the PC space is OS support. Put Win8/Pro and Mac OS in an ARM, now what?

But sooner or later something will happen, and I like that. Its not boring anymore.

In a few years, maybe in 2014, will see Intel vs Nvidia vs AMD vs Samsung vs Qualcomm vs Apple vs Whatever else. Everyone is doing "cpu" and "gpu" now. We will get better products with all that competition. I think.
Click to expand...

Microsoft really needs to lift the restriction on desktop apps for WinRT. On xda there are people running putty, 7zip, and bochs no problem on their Surfaces once they managed to get those unsigned executable to run. I also believe you can run .net apps without recompile (at least that was my impression).
 
Homeles said:
Well there's still cost, which you didn't mention. Possibly Intel's greatest weakness is with cost, rather than performance, perf/watt or power.

But if they did win on every front, like you suggested, then Intel wins. Period. I don't think there's any room for alternative hypotheses.

But cost is a major and often overlooked concern.
Click to expand...

Well, IF ( a big one of course ) they do pull a big rabbit out of their hat, they could afford to sell at low prices for a few years untel ARM is dead and buried, and then having regained their monopoly charge as they please.
 
MightyMalus said:
They can't. They are still playing catch-up. ARM chips aren't stopping improving, Tegra 4 was announced and The new SnapDragons also.
And then comes markets. ARM is every where. Intel is not.

Personally, the only reason I see ARM as limited in the PC space is OS support. Put Win8/Pro and Mac OS in an ARM, now what?

But sooner or later something will happen, and I like that. Its not boring anymore.

In a few years, maybe in 2014, will see Intel vs Nvidia vs AMD vs Samsung vs Qualcomm vs Apple vs Whatever else. Everyone is doing "cpu" and "gpu" now. We will get better products with all that competition. I think.
Click to expand...

Well, with Intel you never know. They might secretly have been working on some advanced tech like optical or something else. Wouldn't be too big a surprise if they did.
 
GillyBillyDilly said:
Well, IF ( a big one of course ) they do pull a big rabbit out of their hat, they could afford to sell at low prices for a few years untel ARM is dead and buried, and then having regained their monopoly charge as they please.
Click to expand...

The EU and US would be all over them for anticompetitive practices.
 
IntelUser2000 said:
Yep, just saw that. Chip companies rarely do this, they are probably really desperate.
Click to expand...

Yea, I am not entirely sure what they are doing to be honest. While I think from a product perspective the roadmap is way better than what has been recently rumored, I am surprised Jaguar is going to be A4/A6-branded with Richland taking up A8 and A10.

Given their perennial inventory issues, my guess would have to be that Kaveri would come up even higher? Maybe they'll turn it up to A11 😎?


Assuming Jaguar actually brings the performance, selling it as a mainstream chip is a clear win for AMD and OEMs. Money all around. But Richland seems much too short-lived to be worthwhile.
 
Homeles said:
Well there's still cost, which you didn't mention. Possibly Intel's greatest weakness is with cost, rather than performance, perf/watt or power.

But if they did win on every front, like you suggested, then Intel wins. Period. I don't think there's any room for alternative hypotheses.

But cost is a major and often overlooked concern.
Click to expand...

Thats pretty bad thinking . The cheapest haswell intel will produce will be the 2 core SoC SOIX chip at 7 watts . It carries the GT3 graphics. There is no law saying intel can not produce a 2 core lowpower chip that meets the requirement of that sector. Intel can sell the SOC haswell at any price over cost . and its a money maker in volumn sales . By going 2 cores and SoS along with SOIX intel allready showed it hand at IDF. This cheap was created for the TABLET market. Its not an accident . Intel can easily charge $10 dollars over its best comp and have the oems eating out of their hand. and make a lot of money doing . The Best part intel did it in plain site . They showed haswell running gt3 8 watts months ago . Talking about blinders
 
IntelUser2000 said:
Yep, just saw that. Chip companies rarely do this, they are probably really desperate.
Click to expand...

Well, Kaveri was suppose to be a 2013 product right? There was no Richland before. Wouldn't AMD be supporting the APU's better if they put the FX chips on hold?

podspi said:
But Richland seems much too short-lived to be worthwhile.
Click to expand...

I hope they don't put much emphasis on Richland. Two or three chips total. They should focus on "next gen" steamroller and excavator AND consumer products.
 
Richland now covers only A8 and A10 lines so AMD is not focusing on it much. It will be decently faster than Trinity in mobile sector(according to pcmar kand 3dmar11 benchmarks AMD gave). Kabini and Temash will cover the A4/A6 and below.
Richland needs to fill in the gap for only 1-2 quarters and then Kaveri will be launched in mobile sector,with much bigger jumps in CPU and GPU performance.
 
IntelUser2000 said:
Yep, just saw that. Chip companies rarely do this, they are probably really desperate.
Click to expand...

I'm not sure what's desperate about it, tbh. Richland is just a respin/deep bin sort of Trinity, as far as I can tell. They have only mentioned mobile chips, so I think they're just finding Trinity dies which can run at higher clocks at the same TDPs as their current range. Basically the same as what Intel is doing with their 13W i3s, except aiming for performance instead of power reduction (Kabini and Temash will cover the low power segment).
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top