You are currently viewing our boards as a guest, which gives you limited access to view most discussions and access our other features. By joining our free community, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content, and access many other special features. Registration is fast, simple, and absolutely free, so please, <a href="/profile.php?mode=register">join our community today</a>!
Post subject: The first "Sandy Bridge" tape-out revealed?
Posted: Sun Jul 05, 2009 10:40 am
Joined: Thu Mar 13, 2008 10:53 am Posts: 15
Quote:
While we can not reveal to you some details of the architecture in the heart of Sandy Bridge, a few remain unknown. First, the Socket. According to the noises heard in corridors at a manufacturer of motherboard, the CPU should resume Socket 1156 of Clarkdale (generation "Westmere" above) but require a new chipset. Then, the performance of the integrated circuit graphics are all subject to speculation. We already know that the IGP Sandy Bridge will always be based on GMA and not on any variation of Larabee. With a frequency of approximately 1.2 GHz, this heart intended to be compatible with DirectX 11, should not be ridiculous but be a lightning war. Finally, the number of PCI Express link is still unknown. According to the first draft of the datasheet, this heart contain 20 lines PCI Express 2.0 (16x + 4x), but the information could not be confirmed ...
Hmm, that 256 Byte (bit ?) ring-bus for the L3 is puzzling me. Does that mean that each CPU core got its own chunk of the L3 array that loops the access requests with the others?!
Very Itanium-like cache organization...
Joined: Wed Aug 29, 2007 3:55 pm Posts: 829 Location: Great white north
jack wrote:
What's the point of putting a relatively large and slow integrated GPU to the already quite large (and therefore not cheap) die?
The integrated GPU enjoys state of the art process technology and
direct access to the CPUs and memory controller. No doubt slower
than a medium and high end discrete GPU it should still be many
times faster than existing chipset based integrated graphics. That
won't satisfy gaming enthusiasts but it will be more than enough
for >90% of the market. It will also be a big win for laptops.
Joined: Wed Jun 27, 2007 10:19 am Posts: 331 Location: Milano, Italy
Paul DeMone wrote:
The integrated GPU enjoys state of the art process technology and direct access to the CPUs and memory controller. No doubt slower than a medium and high end discrete GPU it should still be many times faster than existing chipset based integrated graphics.
You are joking right? Are you aware that current Intel IGPs are many times slower than the poorer offerings of the competition right? How would they magically become 'many times faster' is beyond my imagination. And that's not even mentioning the fact that their drivers are _still_ a joke.
Quote:
That won't satisfy gaming enthusiasts but it will be more than enough for >90% of the market. It will also be a big win for laptops.
A dumb frame-buffer would be more than enough for >90% of the market and a big win for laptops too (can't think of something lower power than that).
Joined: Wed Aug 29, 2007 3:55 pm Posts: 829 Location: Great white north
Gabriele Svelto wrote:
Paul DeMone wrote:
The integrated GPU enjoys state of the art process technology and direct access to the CPUs and memory controller. No doubt slower than a medium and high end discrete GPU it should still be many times faster than existing chipset based integrated graphics.
You are joking right? Are you aware that current Intel IGPs are many times slower than the poorer offerings of the competition right? How would they magically become 'many times faster' is beyond my imagination. And that's not even mentioning the fact that their drivers are _still_ a joke.
I was referring to existing Intel IGP.
Have a look at market share numbers some time, everyone else is
basically noise.
What's the point of putting a relatively large and slow integrated GPU to the already quite large (and therefore not cheap) die?
The integrated GPU enjoys state of the art process technology and direct access to the CPUs and memory controller. No doubt slower than a medium and high end discrete GPU it should still be many times faster than existing chipset based integrated graphics. That won't satisfy gaming enthusiasts but it will be more than enough for >90% of the market. It will also be a big win for laptops.
Putting GPU on die would make sense for dual core part. However putting >50mm^2 GPU on already relatively large quad core part seems to be wasteful, since many buyers of a quad core part want more powerful graphics. Intel could just made a quad core part without GPU and then create CPU+GPU MCM for those who need it.
Joined: Tue Aug 07, 2007 2:04 pm Posts: 33 Location: Tampere, Finland
Gabriele Svelto wrote:
A dumb frame-buffer would be more than enough for >90% of the market and a big win for laptops too (can't think of something lower power than that).
Having only dumb framebuffer would mean all font rendering and bitblits of the GUI would have to be done CPU, and it's slower and more power-consuming to do them on CPU than some dedicated hardware.
Try to run windows XP without graphics drivers, just with "super vga graphics controller" and you will see how much laggier the UI is.
And Vista which renders everything with 3d acceleration would be even much more slower if all that would have to be done on software.
The integrated GPU enjoys state of the art process technology and direct access to the CPUs and memory controller. No doubt slower than a medium and high end discrete GPU it should still be many times faster than existing chipset based integrated graphics. That won't satisfy gaming enthusiasts but it will be more than enough for >90% of the market. It will also be a big win for laptops.
Yeah, the rapidly-expanding segment of the laptop market where 85W spent on the CPU/GPU is commonplace.
1. What makes people think by the time Sandy Bridge is out, Quad Cores will be top performing CPUs? First desktop 6-core CPUs will be out and will become the "new" quad-core.
Quote:
Yeah, the rapidly-expanding segment of the laptop market where 85W spent on the CPU/GPU is commonplace.
I'm pretttyyy sure that the laptop version of the quad-core will be much less than 85W.[/quote]
Joined: Wed Aug 29, 2007 3:55 pm Posts: 829 Location: Great white north
slacker wrote:
Yeah, the rapidly-expanding segment of the laptop market where 85W spent on the CPU/GPU is commonplace.
Yeah, clearly Intel cannot possibly offer substantially lower power
SKUs of this device. Binning by leakage and power scaling with
frequency and voltage doesn't exist. Laptop x86 chips are entirely
distinct and unique designs.
1. What makes people think by the time Sandy Bridge is out, Quad Cores will be top performing CPUs? First desktop 6-core CPUs will be out and will become the "new" quad-core.
Hex is best for the server environment not the Desktop.
Sure it looks like Intel is going to foist Hex core on Desktop suckers, but they supposedly start delivering better targetted to the desktop products starting with Quad Sandy Bridge.
Joined: Wed Jun 27, 2007 10:19 am Posts: 331 Location: Milano, Italy
Paul DeMone wrote:
I was referring to existing Intel IGP.
I see, improvement from their current offerings is largely possible.
Quote:
Have a look at market share numbers some time, everyone else is basically noise.
Being the market leader of a segment with minimal (and ever-shrinking) margins is not exactly a good position to be in. To be fair falling ASPs are not only Intel's problem and they are going to plague pretty much every player in the discrete computer arena (including AMD and nVidia naturally).
Yeah, clearly Intel cannot possibly offer substantially lower power SKUs of this device. Binning by leakage and power scaling with frequency and voltage doesn't exist. Laptop x86 chips are entirely distinct and unique designs.
Scaling to what? (35W Merom/65W Conroe)*(85W Westmere) = 46W? That's horrible. High-end 15" laptops with discrete GPUs and DRAM use about that much power for the entire system at full bore (a quick measurement pegs my laptop at 43W out of the battery under heavy load). Westmere neither offers best performance, nor minimal power consumption, and the die size suggests that it won't be cheap, either.
Users browsing this forum: No registered users and 0 guests
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot post attachments in this forum
