Core i7-3720QM: Ivy Bridge Makes Its Mark On Mobility

[acku]

We've already seen that Ivy Bridge doesn't make much of a splash in the desktop space. But we collected notebooks based on Ivy Bridge, Sandy Bridge, Arrandale, and AMD's mobile Llano, and found that the new architecture's effect on mobility is profound.

Core i7-3720QM: Ivy Bridge Makes Its Mark On Mobility : Read more

 

[fstrthnu]

Would there be a noticeable performance gap between the i7-3720QM and the i7-3612/5QM? I'm trying to decide whether the extra 300 Mhz is worth ~$150 more (which I'm guessing not really)

 

[s3anister]

[citation][nom]fstrthnu[/nom]Would there be a noticeable performance gap between the i7-3720QM and the i7-3612/5QM? I'm trying to decide whether the extra 300 Mhz is worth ~$150 more (which I'm guessing not really)[/citation]

There would be a performance difference in applications that could use the extra MHz (Video games, encoding/decoding) and performance would scale accordingly. Otherwise no you'd likely never notice.

 

[dragonsqrrl]

Wow, it looks like Ivy Bridge is a very compelling option in the mobile market. I had no idea the mobile versions of Llano were so performance constrained by their TDP. The graphics performance results are especially interesting. Just turned my whole world view upside down.

Great job. Another excellent review Andrew.

 

[fstrthnu]

It looks like the Geforce GT650M in the notebook I'm looking at would bottleneck faster than the processor would, so I guess I'll save $$ then

 

[amiame]

A high end desktop plus an ivy bridge ultrabook. Now, THAT works pretty well.

 

[DjEaZy]

... why there waz no screenshots of picture quality differences in games between intel's HD4000 and AMD's HD6620?

 

[blazorthon]

[citation][nom]dragonsqrrl[/nom]Wow, it looks like Ivy Bridge is a very compelling option in the mobile market. I had no idea the mobile versions of Llano were so performance constrained by their TDP. The graphics performance results are especially interesting. Just turned my whole world view upside down.Great job. Another excellent review Andrew.[/citation]

To be fair, it was a low power APU being bench-marked against higher end, higher power, and newer chips. I would be surprised if it won much of anything, besides power usage, against the Sandy and Ivy i7s. A higher TDP mobile A8 might be able to beat HD 4000 if it had 1600MHz or maybe even 1866MHz memory, granted it still wouldn't win in CPU performance.

 

[ojas]

Interesting review. But i guess people are likely to point out differences in price (thus affecting performance/$), and RAM speeds, which apparently impact IGP performance.

IIRC, the IGPs on the mobile chips can be OC'd, right?

The Core i7-3720QM particularly shines in tests involving:

Video Transcoding
DX9 Graphics
Web Browsing

Hmmm...wouldn't you agree that "data decryption" should be on this list too? The difference b/w each proc is significant...plus you've got hardware acceleration for AES256 on SB and IB...

I hope the mobile i3s get HD4000...still wondering why the i5s didn't get it...

 

[DavidC1]

Andrew, love the review. But there's what seems to be a big error. You said on the power usage tests that AMD defaults to max battery life while Intel goes to balanced? Looking at World of Warcraft results, it looks like all the other results may be running max battery life mode for the AMD A8 chip.

The i5-460M is faster than A8-3520M, just not that much faster. I have a feeling you need to run the application and gaming tests on max performance all over again. It doesn't matter for the Intel part as Balanced pretty much performs like max performance.

 

[DavidC1]

Also, you need to do a battery life test. Power usage and battery life tests are hard to connect, because of advanced power management techniques and different usage models.

 

[crisan_tiberiu]

@ 400$ for this benchmarked processor...really, it should beat every Liano ^-

 

[DavidC1]

[citation][nom]ojas[/nom]Interesting review. But i guess people are likely to point out differences in price (thus affecting performance/$), and RAM speeds, which apparently impact IGP performance.IIRC, the IGPs on the mobile chips can be OC'd, right?Hmmm...wouldn't you agree that "data decryption" should be on this list too? The difference b/w each proc is significant...plus you've got hardware acceleration for AES256 on SB and IB...I hope the mobile i3s get HD4000...still wondering why the i5s didn't get it...[/citation]

ojas, All the Core branded mobile chips have the full graphics. For Sandy Bridge that's HD 3000, and for Ivy Bridge its HD 4000. I think you are too much into desktops.

 

[blazorthon]

[citation][nom]ojas[/nom]Interesting review. But i guess people are likely to point out differences in price (thus affecting performance/$), and RAM speeds, which apparently impact IGP performance.IIRC, the IGPs on the mobile chips can be OC'd, right?Hmmm...wouldn't you agree that "data decryption" should be on this list too? The difference b/w each proc is significant...plus you've got hardware acceleration for AES256 on SB and IB...I hope the mobile i3s get HD4000...still wondering why the i5s didn't get it...[/citation]

I think that it's just the desktop i5s and i3s that won't have HD 4000. The mobile ones should have it, kinda like how the mobile Sandy i3s, i5s, and i7s more or less all have HD 3000, but the same is not true for their desktop counterparts. Well, the i5-3570K gets HD 4000, so it's the only exception to the desktop i5s not having HD 4000 and that's just because it's a K edition.

 

[blazorthon]

Honestly I find these benches a bit contradicting compared to the benches done by Anand with the A8 mobiles.

http://www.anandtech.com/show/4444/amd-llano-notebook-review-a-series-fusion-apu-a8-3500m/11

The Review clearly show the A8 with IGP is at least 2X as fast as the HD3000 with i5 CPU

Actually, that clearly shows that the 6620G of the A8s is only marginally better than the HD 3000 and that with the 6630M, the A8s are then closer to twice as fast (some of the time) as the HD3000-using equipped Sandy systems. Keep in mind that the mobile versions of Intel's IGPs are similar to the desktop versions, but the mobile Llano IGPs are much slower than the desktop versions, so on the mobile side, they clash much more, instead of Llano wiping the floor with Intel's IGPs. Trinity will almost certainly let AMD take the lead in mobile graphics IGPs again. Until then, AMD always has the ability to do CF with the IGP and still use similar amounts or even less power than Intel while beating Intel for graphics performance, although Llano clearly can't touch Sandy and Ivy in CPU performance.

 

[jrharbort]

The chart on the first page has a small error. The 3610QM is 2.3GHz, and the 3612QM is 2.1GHz. It's easy to see how the model numbers would be confusing though. The price for both, including the 3615QM, is $378.

http://ark.intel.com/products/family/65506

 

[humbi83]

[citation][nom]s3anister[/nom]There would be a performance difference in applications that could use the extra MHz (Video games, encoding/decoding) and performance would scale accordingly. Otherwise no you'd likely never notice.[/citation]

And don't forget VT-d. That will help you if you are interested in virtualization.

 

[kyuuketsuki]

[citation][nom]blazorthon[/nom]Actually, that clearly shows that the 6620G of the A8s is only marginally better than the HD 3000[...][/citation]
You must not be reading the same article as the rest of us if that's your conclusion.

 

[ojas]

[citation][nom]DavidC1[/nom]ojas, All the Core branded mobile chips have the full graphics. For Sandy Bridge that's HD 3000, and for Ivy Bridge its HD 4000. I think you are too much into desktops.[/citation]
Um...well, i'll cite this statement from Chris Angelini's desktop IB review:

This time around, Intel divides up 3D alacrity a little differently. All mobile and desktop Core i7s get HD Graphics 4000, and all but one (Core i5-3570K) mobile and desktop Core i5s get HD Graphics 2500.

 

[blazorthon]

[citation][nom]Kyuuketsuki[/nom]You must not be reading the same article as the rest of us if that's your conclusion.[/citation]

First benchmark is Battlefield: Bad Company 2
DX10 Low, FRAPS Runthrough

1366x768
A8-3500M+6620G = 48.1FPS
i5-2520M+HD3000 = 30.4FPS
The A8 is 58% faster than the i5.


Second benchmark is Civilization V
DX10/11 Low, LateGameView Benchmark

1366x768
A8-3500M+6620G = 28.6FPS
i5-2520M+HD3000 = 10.7FPS
The A8 is 167% faster than the i5.


Third benchmark is DiRt 2
DX9 Ultra Low, Built-In Benchmark

1366x768
A8-3500M+6620G = 68.1FPS
i5-2520M+HD3000 = 44.3FPS
A8 is 54% faster than the i5.


Fourth benchmark is Left For Dead 2
Low, Timedemo

1366x768
A8-3500M+6620G = 67FPS
i5-2520M+HD3000 = 48.5FPS
A8 is 38% faster than the i5

Fifth benchmark is Mafia 2
Low, Built-In Benchmark

1366x768
A8-3500M+6620G = 34.2FPS
i5-2520M+HD3000 = 16.5FPS
A8 is 108% faster than i5.


Sixth benchmark is Mass Effect 2
Low, FRAPS Runthrough

1366x768
A8-3500M+6620G = 52.1FPS
i5-2520M+HD3000 = 35.8FPS
A8 is 43% faster than i5.


Seventh benchmark is Metro 2033
DX10 Low, Built-In Benchmark

1366x768
A83500M+6620G = 28.6FPS
i5-2520M+HD3000 = 17FPS
A8 is 68% faster than i5.

Eighth benchmark is STALKER: Call of Pripyat
Low + Object, Standalone Benchmark

1366x768
A8-3500M+6620G = 61.7FPS
i5-2520M+HD3000 = 36.3FPS
A8 is 70% faster than the i5.


StarCraft II: Wings of Liberty
Low, FRAPS Playback

1366x768
A8-3500M+6620G = 49.4FPS
i5-2520M+HD3000 = 51.2FPS
i5 is 5% faster than A8.


Ninth benchmark is Total War: Shogun 2
Low, Replay Benchmark

1366x768
A8-3500M+6620G = 79FPS
i5-2520M+HD3000 = 55.1FPS
A8 is 43% faster than i5.

Total for A8-3500M+6620G = 516.8
Total for i5-2520M+HD3000 = 345.8

The A8 is clearly not double the i5. When I said marginally, I missed that the post I replied to referred to i5s, not i7s, so yes, I was wrong on that. However, the post that I replied to was still wrong as well. The i7 would have changed that total FPS from 345.8 to 397.9. The A8 is only 49% faster than the i5, on average, and that is very far from double. The i7 would have brought that down to a mere 30%. Still, I suppose that this is a good deal more than marginally greater, so yes, I was wrong, but nonetheless, so was the post that I replied to. The A8 might be about twice as fast as the mobile i3s with HD 3000, but not the i5s and not even close at that.

 

[Guest]

Why did you only pick out the low quality benchmarks to make your point, blazorthorn? Anantech said that the difference was more pronounced (76% improvement over HD3000) at medium quality.

 

[blazorthon]

[citation][nom]ColinP[/nom]Why did you only pick out the low quality benchmarks to make your point, blazorthorn? Anantech said that the difference was more pronounced (76% improvement over HD3000) at medium quality.[/citation]

I picked the first few benchmarks because I didn't want to do the math with more than ten benchmarks. If you want to, then you go ahead and do it. Either way, 76% is not double and is still a far ways off from it. If it were at least 85% to 90%, then I can understand calling it roughly double. However, not before that.

 

[doron]

The chip is great and everything but.. 378$?! U KIDDING ME?! Intel, you want Ultrabooks to get to the 600-700 dollars range, and if the rest of the lineup is 200+ dollras then I'm afraid manufacturers will have to cut too many corners to hit this target.

 

[blazorthon]

[citation][nom]doron[/nom]The chip is great and everything but.. 378$?! U KIDDING ME?! Intel, you want Ultrabooks to get to the 600-700 dollars range, and if the rest of the lineup is 200+ dollras then I'm afraid manufacturers will have to cut too many corners to hit this target.[/citation]

Keep in mind that it's an i7 and even more, a well binned i7. Ultrabooks will have much cheaper processors available.

 

[doron]

[citation][nom]blazorthon[/nom]Keep in mind that it's an i7 and even more, a well binned i7. Ultrabooks will have much cheaper processors available.[/citation]

Yeah I'm aware of their i7 tax (not bashing, seems completely justified). But since notebook OEMs are saying that Intel's cpu prices are too high, and seeing that a mobile i7 2760QM was also 378$ (pretty comparable) makes me lose hope for a relatively cheap and high-quality Ultrabook for the time being, unless they put this high-quality plastic initiative on high(er) gear.