Intel Announces 'Sandy Bridge' Core Processors

[jb510]

Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_1 like Mac OS X; en-us) AppleWebKit/532.9 (KHTML, like Gecko) Version/4.0.5 Mobile/8B117 Safari/6531.22.7)

I hope Apple offers a configuration that works for me. Essentially I want a 15" MacBook Air Pro. I've long been willing to ditch the optical drive in my MBP and would love an SSD, but only if tree is USB3 for a big fat and fast external for my photo and iTunes library (things I don't need 24/7 access to but are way to big to be in the cloud)... oh... And hybrid graphics i still wouldnt trust IGP by itself, especially when I hook up a 24" external LCD.

Now if it still has a hard drive, or has hybrid SSD/HDD I'm cool with that... Still want USB 3. If it stillmhas an optical drive I'll be annoyed but again I can accept it... No USB 3 or no dedicated graphics though and it's another year of waiting to upgrade for me.

 

[KnightWRX]

They just started the whole GPU stuff a few years ago. You cannot do this from one day to the next.

Wasn't the GMA 950 released like 5 years ago? The fact is every IGP release has seen very dismal performance gains. The sandy bridge IGP is a prime example, a IGP that barely tries to stack up against a IGP that was released 2 years ago.

The poster you responded to is making stuff up and it shows. Intel has been in the graphics game since at least 1998, when they released the i740. It was hyped as the "fastests 3D accelerator yet! Built with AGP support from the get-go!". It was abysmal.

13 years later, and they are still 2 years behind their competition. nVidia has been shipping graphics processor since the NV-1 in 1995 and were behind 3Dfx until at least the Riva TNT chipset, which closed the gap, a mere 3 years after starting to make graphics.

And let's talk about 3Dfx. They shipped their first product in 1996. They were instantly crowned king, having the best products and holding that position for quite a while. Not bad for someone who has been at it for "a short time" and just went for nobodies to the best in "one day to the next".

Intel has gobs of money and engineering talent. At this point, there is no excuse for them for their paltry showing in graphics technology. They aren't investing what they need to be into it.

 

[Richard1028]

SB Sucks...

I don't understand the giddy comments in this thread. Particularly those interested in a new MPB13.

When benchmarked against the current MBP13 (C2D+320M)... Sandy Bridge and its IGP lost out in 6 out of 7 gaming tests.

And that SB processor was a top of the line i7 mobile chip! (2820QM). God only knows what will go into a new MBP13 but it ain't gonna be an i7.

I'm obviously missing something. Somebody please skool me as to why this is such a celebratory milestone when it comes to the 13" MBP.

 

[Env]

I'm totally lost here too. Iv read the whole thread, so is it worth sticking around for these sandy bridge processors, or get what's on offer now and maybe upgrade again at ivy bridge, gonna be doing some fairly heavy audio editing?

 

[KPOM]

*sigh*
In fact, it is NOT a new architecture.
It's the same marketing jamboowamboo when the Core 2 CPUs were introduced, which weren't a new architecture either.

Just like Merom was a refinement of its older Pentium M (Yonah) sibling, Sandy Bridge is a refinement of its Nehalem predecessor. Pentium 4 WAS an entirely new architecture at its time, but it was abandoned.

Merom/Core 2 was actually Yonah+x64. Pentium M chips were 32-bit only. Yonah itself was modified Pentium III. Basically, Intel realized that Apple/IBM/Motorola were right about CPU performance being more than clock speed, and abandoned the Pentium 4 platform.

 

[KPOM]

Intel has gobs of money and engineering talent. At this point, there is no excuse for them for their paltry showing in graphics technology. They aren't investing what they need to be into it.

Perhaps they just don't care. Their older chips were "good enough" for those in the sub-$500 notebook market, and now it appears that Sandy Bridge will be "good enough" for MacBook and MacBook Air users.

 

[Maven1975]

Honestly, any Mac without a dedicated GPU should not carry the Pro name. Whats the point of the Pro name without it?

 

[Eidorian]

Honestly, any Mac without a dedicated GPU should not carry the Pro name. Whats the point of the Pro name without it?

Your money.

Perhaps they just don't care. Their older chips were "good enough" for those in the sub-$500 notebook market, and now it appears that Sandy Bridge will be "good enough" for MacBook and MacBook Air users.

Apple can just change to propaganda to show the improvement on the CPU side and simply avoid doing a comparison for the IGP. It still manages to keep pace with the GeForce 320M and the near vaporware status of OpenCL need not apply.

 

[commander.data]

You're mixing layers here. AVX is, very broadly speaking, a vector processing unit. OpenCL is a software layer and will be able to target that too as a compute resource. I also think Intel will bring an IGP with OpenCL/DirectX 11 support at some point. It has to to bridge the gap between now and the commoditization of the Larrabee-derived Knights Corner/Ferry.

I've not said anywhere that Sandy Bridge is of no use, or that AVX doesn't have some significance. My point is that Apple do focus on OpenCL and that has some bearing on when, and if, they adopt a hardware choice that doesn't accommodate that well.

I was just taking the point of view that AVX and OpenCL are broadly reaching for the same middle group, where AVX is trying to make CPUs more GPU-like in vector processing capabilities whereas OpenCL is trying to making GPUs more CPU-like in ability to target non-graphics workload. Of course OpenCL also encompasses CPUs. I'm thinking that if AVX and OpenCL are both focused on providing enhanced vector processing capabilities to software, but in the near term AVX will see faster, broader adoption since companies can migrate their existing SSE codebase rather than developing a new OpenCL codebase, then it makes sense to push AVX first. For instance in a hypothetical component selection runoff between Sandy Bridge and AMD's Llano which has a Core 2 Duo level CPU and no AVX support although it has a fast IGP with OpenCL support, it probably makes sense to pick Sandy Bridge since most software can readily take advantage of the faster CPU and AVX rather than a few software than can take advantage of OpenCL.

Wasn't the GMA 950 released like 5 years ago? The fact is every IGP release has seen very dismal performance gains. The sandy bridge IGP is a prime example, a IGP that barely tries to stack up against a IGP that was released 2 years ago.

Who really wants to play games at low settings? There's a lot of games you can play at medium with the 320m yet the sandy bridge IGP ALTHOUGH paired with a faster CPU is 26% slower than a 320m at medium settings? Screw intel and their lack of focus in their GPU sector.

If you are saying every Intel IGP release has seen dismal performance gains, that really isn't true. Sandy Bridge doubles the IGP performance of Arrandale and Arrandale was pretty much 2 times faster than the previous HD4500. When measuring relative performance increases, Intel has been achieving consistent doubling of graphics performance which nVidia and AMD have not been able to achieve. The GTX580 is not 2 times faster than the GTX480, it's not even 50% faster. Neither is the HD6970 2 times faster than the HD5870. Admittedly, Intel isn't pushing the performance envelope. Although since the 320M is less than 1 year old, Intel isn't meeting a 2 year gap as you say.

In terms of medium settings, from Anand's benchmarks, even though the 320M beats the HD 3000, it might not be that relevant since the 320M floats dangerously around the 30fps threshold, which is often considered as playable. So the 320M may not provide a worthwhile playing experience anyways at medium settings. It's too bad Anand didn't include the 320M in their extended game tests for more data points.

Honestly, any Mac without a dedicated GPU should not carry the Pro name. Whats the point of the Pro name without it?

Full voltage processors and aluminum seems to be the minimum requirement for Pro. Now if Pro meant IPS screens then that could make it worthwhile even without a dedicated GPU. Going IPS would probably cost more than a dedicated low-end GPU too.

 

[KnightWRX]

Perhaps they just don't care. Their older chips were "good enough" for those in the sub-$500 notebook market, and now it appears that Sandy Bridge will be "good enough" for MacBook and MacBook Air users.

The only thing their chips are "good enough" at is holding back the PC graphics industry. If Intel had shipped nVidia quality IGPs since day 1, gaming and other intensive graphical applications would be much more common on PCs these days.

Although since the 320M is less than 1 year old, Intel isn't meeting a 2 year gap as you say.

We're in January 2011. The first iMac with a 320M appeared in October 2009. The 320M's IGP is from a family of graphics that appeared before that (look for the 310M).

Your numbers are off.

 

[Richard1028]

It's all smoke and mirrors...

Apple better not skip on the dedicated GPU in the 13 refresh!

We're screwed.

I'm convinced there will be no discrete graphics in the 13" MBP. Intel is applying tons of lipstick on the Sandy Bridge IGP and Apple will spin it even further.

By the time Steve talks about it the new MBP13 will even wipe your arse.

"Bend over... Boom! There, you're done!"

 

[lilo777]

I don't understand the giddy comments in this thread. Particularly those interested in a new MPB13.

When benchmarked against the current MBP13 (C2D+320M)... Sandy Bridge and its IGP lost out in 6 out of 7 gaming tests.

And that SB processor was a top of the line i7 mobile chip! (2820QM). God only knows what will go into a new MBP13 but it ain't gonna be an i7.

I'm obviously missing something. Somebody please skool me as to why this is such a celebratory milestone when it comes to the 13" MBP.

And who is buying Mac computers for gaming? They all are hopelessly behind even much cheaper PC counterparts in this department. SB is almost ideal for MBP because of what they are - internet browsing and office task machines.

 

[Eidorian]

We're in January 2011. The first iMac with a 320M appeared in October 2009. The 320M's IGP is from a family of graphics that appeared before that (look for the 310M).

Your numbers are off.

The GeForce 310M is actually a higher clocked GeForce 210M... The GT216 core is from 2009 and a scaled down GT200. Ivy Bridge later this year plans on doubling the number of execution units for 24 for the IGP and dedicated VRAM has been rumored as well.

MCP89 was at the very least 6 months late when it showed up in Spring 2010. I was talking about it back in 2009 along with MCP99.

 

[brentsg]

Intel is applying tons of lipstick on the Sandy Bridge IGP and Apple will spin it even further.

By the time Steve talks about it the new MBP13 will even wipe your arse.

"Bend over... Boom! There, you're done!"

Using past Intel graphics solutions as a guide, arse wiping would be an appropriate utilization for the part.

 

[commander.data]

We're in January 2011. The first iMac with a 320M appeared in October 2009. The 320M's IGP is from a family of graphics that appeared before that (look for the 310M).

Your numbers are off.

Was there ever an iMac with the 320M? I believe the iMac thankfully jumped from the 9400M directly to the HD4670, which although last gen is still pretty aggressive GPU-wise from Apple. I thought the 320M launched with the 13.3" MacBook Pro in April 2010?

 

[tirerim]

Wirelessly posted (iPhone 3GS: Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_2_1 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8C148a Safari/6533.18.5)



I agree with you to a point. I wish we (in America) would deploy the kind of broadband speeds they have in Europe. I was blown away at the speeds they have up and down on another MR thread about posting your speed test results. I thought the fiber connection I have at work was fast, but compared to some in the old country, I may as well have been using a dialup modem.

But Intel is doing what they do, develop processors, not hard drives or SSD's. As far as networking and bandwidth, it will take the whole industry to upgrade forklift style. Think of all the different router makers, switches, NIC's, etc... out there that need to start moving forward. I agree with you though, I wish we'd start moving faster on the networking side. Storage speed is finally getting a boost with the SSD market.

Yeah, the problem in the States is infrastructure: our population density is much lower than in Europe, so there's a lot more fiber to deploy (and all the other associated hardware) before a new technology can get decent penetration. I'm guessing that in at least some parts of Europe the infrastructure deployment is at least partly government-supported, too; here, we haven't been willing to do that sort of thing since rural electrification in the early 20th century. There's some hope that heavily urban areas might achieve decent speeds in the not-too-distant future (Google is planning an experiment along those lines), but it's unlikely we'll ever catch up in general.

 

[applefanDrew]

Was there ever an iMac with the 320M? I believe the iMac thankfully jumped from the 9400M directly to the HD4670, which although last gen is still pretty aggressive GPU-wise from Apple. I thought the 320M launched with the 13.3" MacBook Pro in April 2010?

Agreed...I thought the 320m was made specifically for the 13 MBP. I read it in a press release when it came out or something.

 

[KnightWRX]

I thought the 320M launched with the 13.3" MacBook Pro in April 2010?

Looking through the net for archives, yes, it does seem the 21.5" iMac shipped with a 9400m in October 2009. The 300M series is from January 2010 and the actual 320M first appeared in Apple computers in April of 2010.

Doesn't make the Intel numbers any more impressive after 13 years in the game, but nonetheless, it's not as bad as "2 years behind", they're just 1 year behind now.

 

[Hellhammer]

Ivy Bridge later this year plans on doubling the number of execution units for 24 for the IGP and dedicated VRAM has been rumored as well.

Anand is saying that IB will only add 4 EUs

Unfortunately I’m worried that Ivy Bridge won’t bring another doubling as it only adds 4 EUs to the array.

http://www.anandtech.com/show/4083/...-i5-2600k-i5-2500k-and-core-i3-2100-tested/10

 

[alent1234]

The question is whether AVX and Sandy Bridge's dedicated media encoder is an acceptable alternative to OpenCL running on an nVidia IGP or low-end discrete GPU. The most common use case for GPGPU is video encoding, and Anand's benchmarks have shown that Sandy Bridge's dedicated media encoder is faster than even a GTX465M using CUDA which is more mature than OpenCL. In other words, lack of OpenCL GPGPU acceleration is in no way an impediment for Sandy Bridge's video encoding abilities. As for everything else, especially third-party software, it's much easier for them to transition existing code from SSE acceleration to AVX acceleration for a noticeable speedup than to rewrite code in OpenCL. The AVX speedup may not exceed what OpenCL can accomplish, but if it's easier to realize, I think that will be most beneficial for most users.

EDIT: As an observation, it appears from Anand's review that Sandy Bridge brings noticable improvements in battery life where a quad core Sandy Bridge can get similar minutes of internet usage per MWhr of battery capacity as dual core Arrandale which is a significant improvement over Clarksfield quad cores. More likely than not, battery life concerns will be the primary factor in Apple deciding whether to go pure Sandy Bridge as opposed to Sandy Bridge + discrete GPU or Core 2 Duo + IGP or any other combination rather than GPU performance concerns for space constrained models like the 13.3" units. In other words, much better CPU performance than Penryn Core 2 Duo, equivalent GPU performance as the 320M, and noticeable improvement in battery life seems like a compelling argument even if OpenCL support is quietly relegated to CPU support only. Apple will no doubt focus on the video encoding acceleration performance improvement anyways.

Tom's Hardware said that the video encoding/decoding hardware in SB is at least 1 year ahead of nvidia and AMD. intel had been working on it for over 5 years and only announced it last year.

it might not play games very fast, but it outperforms the competition in tasks most people use computers for

 

[Richard1028]

And who is buying Mac computers for gaming?

Certainly not me and I've admitted as much many times over. But I didn't chose the tests.

SB is almost ideal for MBP because of what they are - internet browsing and office task machines.

Agree 100% but specifically regarding the MBP13...

Going "backwards" in gaming performance is just stupid even for Apple. (what little performance there is of course). When a current MBP13 with an ancient C2D+320M smokes a brand spankin' new QUAD core mobile i7 in the tests, well... that's just plain stupid. No other word for it.

Good for me though. Looks like I'll be getting a new MBP13 C2D real cheap. I'll wait to see if Apple puts discrete graphics along with the SB processor in the MBP13 first.

Gotta go... these flying pigs need to be fed.

 

[LarryC]

Eh, the numbers are easily manipulated. Remember it's all about marketing.

There are three types of lies.
1. Lies
2. Damn Lies
3. Statistics

 

[un.titled]

It's sad that a $499 Acer laptop will have this months before the $2000 MBP.

 

[commander.data]

Certainly not me and I've admitted as much many times over. But I didn't chose the tests.

Agree 100% but specifically regarding the MBP13...

Going "backwards" in gaming performance is just stupid even for Apple. (what little performance there is of course). When a current MBP13 with an ancient C2D+320M smokes a brand spankin' new QUAD core mobile i7 in the tests, well... that's just plain stupid. No other word for it.

Good for me though. Looks like I'll be getting a new MBP13 C2D real cheap. I'll wait to see if Apple puts discrete graphics along with the SB processor in the MBP13 first.

Gotta go... these flying pigs need to be fed.

Maybe I'm mis-interpreting you, but you say gaming isn't a determining factor in buying a Mac and yet you say you'll be getting a C2D+320M MacBook instead of a Sandy Bridge MacBook due to the gaming benchmarks. Since by any other benchmark than gaming, Sandy Bridge looks superior to Nehalem/Westmere much less Penryn. It seems contradictory.

Besides, Sandy Bridge's IGP is looking more like a sidegrade from the 320M. It's faster at low settings, slower at medium settings but even that has a caveat that at medium settings in most of Anand's benchmarks Sandy Bridge and the 320M really tied since the 320M is just ~1fps faster, likely within their error range. And given the performance level of these IGPs, the games are most likely GPU-bound so going from a quad core Sandy Bridge to a dual core Sandy Bridge shouldn't change the results much.

 

[gorcman]

So if I'm reading the benchmarks correctly, if the new 13" MBP gets the Sandy bridge treatment, we won't really see any big increases in gaming performance compared to the current models?

Looking at the benchmarks, Sandy Bridge is using an i7 in these tests, and its gaming performance is more or less on par with the MBP C2D+320M in medium performance.

And somehow i doubt we'll see i7's in the entry level 13" MBP, meaning the performance gains would be less if the 13" MBP received an i3 for example (though it'd be a nice surprise if we did get an i7).)Regardless it looks like Intel's IGP will be the new bottleneck. Any hope we'll see an Nvidia or ATI GPU in the 13"?