• 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.

Intel Q7500 : SLG9U 2.60GHz 4 800MHz 45nm R0 2MB

Wow... not even 2x2MB? That's gotta be starved, just like the Phenoms. Looks interesting with the 13x multiplier but it only has 2MB cache... Anyway, likely a part that will never be seen in the US.
e7500 and e5400 are coming out soon if I'm not mistaken. They'll just have higher multiplier than the already released e7400 and e5300.
 
Originally posted by: Andrew1990
God, I hope it comes at the $100 price mark. Intel would kill the competition with a ~$100 Quad Core
Click to expand...

I agree...if it is like the price points of the e5200 and e7200 then it could be potentially sweet....
 
A $100 quad-core would be interesting to be sure.

I'm concerned about the 2MB cache. I really, really, hope that's some sort of typo.

If they're slapping two E5300 die together to make that quad (R0 stepping), I would hope that they would end up with 4MB of cache.
 
Originally posted by: VirtualLarry
A $100 quad-core would be interesting to be sure.

I'm concerned about the 2MB cache. I really, really, hope that's some sort of typo.

If they're slapping two E5300 die together to make that quad (R0 stepping), I would hope that they would end up with 4MB of cache.
Click to expand...

Yea, only combination I could think of would to be slapping 2x E2xxx chips together to use up their old stock. In that case, this chip would suck. On the other hand, those are all 65nm pieces while this one is 45nm?
 
This chip looks like a couple of neutered E5000 or maybe even E7000 chips. Reduce the cache size and FSB of each die, slap two together, and we have a quad. Performance would be intersting. Seems like it will probably perform worse than most Phenom X4s and maybe a few Phenom X3s, but I don't know how that small cache size will affect the 45nm chips. The 65nm Pentium chips took a huge hit in performance from the reduced cache.

I also doubt these would be around the $100 price mark, although I could see them as low as $125. On the plus, these will probably be good overclockers and would work best for a workstation-type machine. The low power consumption is even more intriguing.
 
Doubt the 65W parts (also one of the q8200s) will be available to public.

Never mind... via link on last post of this old thread, looks like the 3 65w quads (some reason q9400 isn't on the site) officially launched 1/18.
 
Well, the 45nm "Pentium" line (e5x00) features 2MB shared cache (1MB/core). This chip will have 2MB/4 cores = 512k/core. Making this maybe a Celeron Quad Core?

Has anyone seen benchies comparing cache level impact on duals versus quads? I know the quads are more sensitive to the fsb speed (gain more benefit from higher fsb than duals) and I wonder if this would hold true for the amount of cache available.

I'm guessing this chip is gonna suck. The Q8200 looks pretty neutered in most benchies I've seen and this takes the cache to a whole new level of bad.
 
Originally posted by: Denithor
Well, the 45nm "Pentium" line (e5x00) features 2MB shared cache (1MB/core). This chip will have 2MB/4 cores = 512k/core. Making this maybe a Celeron Quad Core?

Has anyone seen benchies comparing cache level impact on duals versus quads? I know the quads are more sensitive to the fsb speed (gain more benefit from higher fsb than duals) and I wonder if this would hold true for the amount of cache available.

I'm guessing this chip is gonna suck. The Q8200 looks pretty neutered in most benchies I've seen and this takes the cache to a whole new level of bad.
Click to expand...

Seriously.

If Intel wants to push a quad-core SKU in name only, with utter lack of performance to go along with it, why not make an MCM'ed quad-core Atom package and be done with it.

Save themselves money and still get to put a sticker on the box that says "unleash new quad-core technology for maximum performances".

What possible purpose other than marketing could there be to having four cores when the IPC for the cores will be total suckage due to lack of cache? I can't think of any apps where this becomes a good product relative to a dual-core with more cache.
 
Originally posted by: Idontcare
Originally posted by: Denithor
Well, the 45nm "Pentium" line (e5x00) features 2MB shared cache (1MB/core). This chip will have 2MB/4 cores = 512k/core. Making this maybe a Celeron Quad Core?

Has anyone seen benchies comparing cache level impact on duals versus quads? I know the quads are more sensitive to the fsb speed (gain more benefit from higher fsb than duals) and I wonder if this would hold true for the amount of cache available.

I'm guessing this chip is gonna suck. The Q8200 looks pretty neutered in most benchies I've seen and this takes the cache to a whole new level of bad.
Click to expand...

Seriously.

If Intel wants to push a quad-core SKU in name only, with utter lack of performance to go along with it, why not make an MCM'ed quad-core Atom package and be done with it.

Save themselves money and still get to put a sticker on the box that says "unleash new quad-core technology for maximum performances".

What possible purpose other than marketing could there be to having four cores when the IPC for the cores will be total suckage due to lack of cache? I can't think of any apps where this becomes a good product relative to a dual-core with more cache.
Click to expand...
To sell more chips. To keep fabs going. It makes sense.
As you said the marketing would fool regular Joe.
 
Originally posted by: Idontcare
Originally posted by: Denithor
Well, the 45nm "Pentium" line (e5x00) features 2MB shared cache (1MB/core). This chip will have 2MB/4 cores = 512k/core. Making this maybe a Celeron Quad Core?

Has anyone seen benchies comparing cache level impact on duals versus quads? I know the quads are more sensitive to the fsb speed (gain more benefit from higher fsb than duals) and I wonder if this would hold true for the amount of cache available.

I'm guessing this chip is gonna suck. The Q8200 looks pretty neutered in most benchies I've seen and this takes the cache to a whole new level of bad.
Click to expand...

Seriously.

If Intel wants to push a quad-core SKU in name only, with utter lack of performance to go along with it, why not make an MCM'ed quad-core Atom package and be done with it.

Save themselves money and still get to put a sticker on the box that says "unleash new quad-core technology for maximum performances".

What possible purpose other than marketing could there be to having four cores when the IPC for the cores will be total suckage due to lack of cache? I can't think of any apps where this becomes a good product relative to a dual-core with more cache.
Click to expand...

Haven't they essentially already with the Atom 330? I know it's really hyperthreading but still..
 
Originally posted by: Denithor
Well, the 45nm "Pentium" line (e5x00) features 2MB shared cache (1MB/core). This chip will have 2MB/4 cores = 512k/core. Making this maybe a Celeron Quad Core?

Has anyone seen benchies comparing cache level impact on duals versus quads? I know the quads are more sensitive to the fsb speed (gain more benefit from higher fsb than duals) and I wonder if this would hold true for the amount of cache available.

I'm guessing this chip is gonna suck. The Q8200 looks pretty neutered in most benchies I've seen and this takes the cache to a whole new level of bad.
Click to expand...

i think it'd probably be a 2 x 2MB l2 cache design. there is no 45nm counterpart with 1MB of cache per 2 cores.
 
Originally posted by: AmongthechosenX
Originally posted by: Denithor
Well, the 45nm "Pentium" line (e5x00) features 2MB shared cache (1MB/core). This chip will have 2MB/4 cores = 512k/core. Making this maybe a Celeron Quad Core?

Has anyone seen benchies comparing cache level impact on duals versus quads? I know the quads are more sensitive to the fsb speed (gain more benefit from higher fsb than duals) and I wonder if this would hold true for the amount of cache available.

I'm guessing this chip is gonna suck. The Q8200 looks pretty neutered in most benchies I've seen and this takes the cache to a whole new level of bad.
Click to expand...

i think it'd probably be a 2 x 2MB l2 cache design. there is no 45nm counterpart with 1MB of cache per 2 cores.
Click to expand...

Unless they intentionally disable cache blocks and restrict the SKU to 2x1MB L2$. Nothing in the speculation to date precludes them from such shenanigans.
 
Well, it's supposedly a <$150 cpu, which could be attractive and swing a lot of AMD Phenom/x3 considerers into Intel camp.
About the 3 65w quads, are they really worth the ~$55 price premium?
 
Originally posted by: Idontcare
Seriously.

If Intel wants to push a quad-core SKU in name only, with utter lack of performance to go along with it, why not make an MCM'ed quad-core Atom package and be done with it.

Save themselves money and still get to put a sticker on the box that says "unleash new quad-core technology for maximum performances".

What possible purpose other than marketing could there be to having four cores when the IPC for the cores will be total suckage due to lack of cache? I can't think of any apps where this becomes a good product relative to a dual-core with more cache.
Click to expand...

I agree. This is going to give Intel quad-cores a BAD name, among the budget buyers. Might even push some of them over to AMD.

I hate it when marketing wins out over engineering. I think that's whats happened here.
 
Originally posted by: Denithor
Well, the 45nm "Pentium" line (e5x00) features 2MB shared cache (1MB/core). This chip will have 2MB/4 cores = 512k/core. Making this maybe a Celeron Quad Core?

Has anyone seen benchies comparing cache level impact on duals versus quads? I know the quads are more sensitive to the fsb speed (gain more benefit from higher fsb than duals) and I wonder if this would hold true for the amount of cache available.

I'm guessing this chip is gonna suck. The Q8200 looks pretty neutered in most benchies I've seen and this takes the cache to a whole new level of bad.
Click to expand...

How is the Q8200 neutered? From ATs own benchmarks in the Phenom II review:
-----------------------------------------------------------------------------
Adobe Photoshop CS4 using Retouch Artists Speed Test
Q8200 - 23.7 [wins]
Q6600 - 25.6

DivX 8.5.3 with Xmpeg 5.0.3
Q8200 - 57.6
Q6600 - 56.4 [wins]

x264 HD Encode (First Pass)
Q8200 - 53.3
Q6600 - 53.8 [wins]

x264 HD Encode (Second Pass)
Q8200 - 15.4 [wins]
Q6600 - 15.2

Windows Media Encoder 9 x64 Advanced Profile
Q8200 - 40
Q6600 - 40

3dsmax 9 - SPECapc 3dsmax CPU Rendering Test
Q8200 - 9.4
Q6600 - 9.6 [wins]

Cinebench R10 Single Threaded
Q8200 - 2788 [wins]
Q6600 - 2778

Cinebench R10 Multi Threaded
Q8200 - 10047 [wins]
Q6600 - 9681

POV-ray 3.7 Beta 23
Q8200 - 1963
Q6600 - 1996 [wins]

Multi-threaded par2 Archive Recovery
Q8200 - 43.2 [wins]
Q6600 - 47.6

Blender 2.48a
Q8200 - 70.1 [wins]
Q6600 - 77.1

Microsoft Excel 2007
Q8200 - 26.2
Q6600 - 22.1 [wins]

Sony Vegas Pro 8: Blu-ray Disc Creation
Q8200 - 317.6 [wins]
Q6600 - 324.4

Sorenson Squeeze: FLV Creation
Q8200 - 180.2
Q6600 - 179 [wins]

WinRAR - Archive Creation
Q8200 - 147.2
Q6600 - 140.7 [wins]

Fallout 3
Q8200 - 73 [wins]
Q6600 - 67.4

Left 4 Dead
Q8200 - 98.8
Q6600 - 101.8 [wins]

FarCry 2
Q8200 - 50.6 [wins]
Q6600 - 48.7

Crysis Warhead
Q8200 - 68.1
Q6600 - 69.8 [wins]
-----------------------------------------------------------------------------

Out of the 19 tests, the Q8200, despite having a lower clock speed and half the L2 cache, beats or ties the Q6600 in 10 tests. I would do a percentage calculation, but I don't think it's necessary for the point I'm making. And across several other review sites I see similar results: The Q8200, despite being neutered, does not perform like it has its balls cut off and is just as fast as the Q6600.

So the Q8200 isn't exactly cached starve. And what is intriguing, and what we won't know until we get some benchmarks, is just how small the L2 cache can go with 45nm process before really suffering. My guess is that the new chip will suffer and be worse than many of the sub-$150 Phenoms, but maybe not.
 
thing is the q9400s is not the same chip as listed before the fsb is 1333 instead of 800 and thus the multiplier is different.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top