NotebookForums.com › Forums › General Notebook Discussions › Notebook Forums - General › Have you all seen these Turion Benchmarks?
New Posts  All Forums:Forum Nav:

Have you all seen these Turion Benchmarks? - Page 4

post #61 of 99
4/27/05 at 12:35am
Quote:
Originally Posted by MichaelX30
Now that I got that out of my system I am gonna so something more productive for the night other than trying to explain to an idiot the difference between 533 MHz FSB and 400 MHz FSB
Hey, if you were so smart, why don't you tell me the intricacies of the 533MHz, vs the 400MHz FSB, how a 2.13GHz CPU can run on both types of system clocks??
Reply
Reply
post #62 of 99
4/27/05 at 12:42am
Thread Starter 
Quote:
Originally Posted by HardBall
I didn't mean that the review had the Alviso or Sonoma, but I was just saying that even if it's with the Alviso, things would not be that different. Sorry if I caused you to misunderstand me.

And please read the rest of my post, I wasn't even finished when you posted.
Of course things would not be that different, after all since it is Intel a FSB clocked 33% higher and Ram clocked 20% higher makes zero difference... Since you've shown me proof that the Alviso lags behind the 3400+ (which you havent, you have only showed Intels last gen platform against it) I guess you can call it a "win"

We'll I'm just gonna give up, I am tired of talking to a person who fails to realize that fast clock speeds ultimately lead to better performance.
Reply
Reply
post #63 of 99
4/27/05 at 12:47am
Quote:
Originally Posted by MichaelX30
No no no no no no no

What about this do you not understand?

Calm down mike, when you get frustrated you call other people names and tend to be harsh

I'll be back in 5 minutes after a cool down

Edit: Ok I'm back

Those benchmarks are also using the Centrino technology which consists of the following

1. The Dothan with it's FSB @ 400 MHz
2. DDR-I Ram @ 333 MHz

The benchmarks that are against the Turion are using SONOMA intels updated platform which include

1. The Dothan with it's FSB @ 533 MHz, a 33% increase
2. DDR-I Ram @ 400 MHz or DDR-II Ram @ 533 MHz (The benches against the Turion used DDR-I @ 400 MHz)

The benchies in the thread you posted are not using Intel's newest platform, how hard is that to understand? How damn complicated is that to grasp? There is no P-M paired with Alviso which allows it to run @ 2.1 GHz like you are claiming.

The Current Alviso/Dothan speeds include

P-M 770, 2.13 GHz/533 MHz FSB
P-M 760, 2.00 GHz/533 MHz FSB
P-M 750, 1.86 GHz/533 MHz FSB
P-M 740, 1.73 GHz/533 MHz FSB
P-M 730, 1.60 GHz/533 MHz FSB

The ones in the benhies you posted are not using Intels latest platform, they are

P-M 765, 2.10 GHz/400 MHz FSB
P-M 755, 2.00 GHz/400 MHz FSB
P-M 745, 1.8 GHz/400 MHz FSB
P-M 735, 1.6 GHz/400 MHz FSB

Do you see the difference now? I really don't have the patience to teach this to you, after all you should know the difference since you are always correct.
I know all that full well, if you have ever read my previous posts you will know that. Such as these:

http://notebookforums.com/showthread.php?t=76011
http://notebookforums.com/showthread...hlight=533+133

I have posted repeated on these boards about the difference between system clock, memory clock, FSB. 2.1 is a common shorthand for 2.13333333333333333 GHz, and everyone in the forum knows that fact. Please don't get so emotional when someone uses a shorthand to indicate the speed of the FSB of your dear old PM, or slighting the CPU clock by one thirtieth of 1GHz.
Reply
Reply
post #64 of 99
4/27/05 at 12:47am
you guys need to chill. The difference between the sonoma and caramel is roughly 5-10%. So even if the benchmark used the latest platform, it still wouldn't have made much of a difference. I agree somewhat with hardball though. The 2.13 ghz p-m would be roughly equivalent to a 3400+/3700+ amd 64. Those benchies are definitely weird and completely opposite of what toms and anand reported.
Reply
Reply
post #65 of 99
4/27/05 at 12:51am
Quote:
Originally Posted by nark
The difference between the sonoma and caramel is roughly 5-10%.
I like to see the bench on that...
Reply
Reply
post #66 of 99
4/27/05 at 12:54am
Mike and Hardball.... You guys take it easy.

Clearly Turion is not a P-M killer but it should be competitive in the portable thin and light segment. Anyway it will be good for the consumer as there will be more choices.
Reply
Reply
post #67 of 99
4/27/05 at 12:56am
Quote:
Originally Posted by MichaelX30
Of course things would not be that different, after all since it is Intel a FSB clocked 33% higher and Ram clocked 20% higher makes zero difference... Since you've shown me proof that the Alviso lags behind the 3400+ (which you havent, you have only showed Intels last gen platform against it) I guess you can call it a "win"

We'll I'm just gonna give up, I am tired of talking to a person who fails to realize that fast clock speeds ultimately lead to better performance.
You have to realize that I did, and that the 3500+ has virtually the identical performance as the 3400+, as one has dual channel, the other 1MB cache, each's strength compensating for its weakness. Both clocked at 2.2 GHz. The 3400+ is usually replaced by the 3500+ because reviewers can simply use the S939 board instead of another S754:

Here is another review:

http://www.pcper.com/article.php?aid...e=expert&pid=7
Reply
Reply
post #68 of 99
4/27/05 at 12:58am
I agree with Nark, and here is the proof

Here is a bench from Tom's using the proper 533 DDR II and ICH6 915 chipset with pm 2.13 that show an average 5-10% increase between old centrino and 2.13 sonoma with 533 memory.
http://www.tomshardware.com/mobile/2...alviso-10.html

So if you take a 10% increase and compare it with other benches, pm should be equal to 3400+, not that much superior.
Reply
Reply
post #69 of 99
4/27/05 at 1:47am
Everybody has to wait for AnandTech to come up with a review, that is the only people I trust, it is kind of hard to believe anyone else. People are too biased. Wait for a real comparison like the Aspire 5024 and the Travelmate 8104 or something, and then judge. Probably no one will see and believe what they read really until one of us has it and show the benchmarks here. That is the way always things work isn't it? One of us gets the laptop and show it off @ NBF and then others can see what all the hype is all about.
Reply
Reply
post #70 of 99
4/27/05 at 1:52am
I want to add, either way, at least engineers are going to be really interested in Turion64 because almost every single engineering software is 64bit ready. It is going to be really interesting to work in AutoCAD or ProE and that the damn software flyes instead of taking their whole life. Engineers need efficient systems and I bet Turion64 would satisfy that.
Reply
Reply
post #71 of 99
4/27/05 at 2:48am
Quote:
Originally Posted by gilgalasfv
I agree with Nark, and here is the proof

Here is a bench from Tom's using the proper 533 DDR II and ICH6 915 chipset with pm 2.13 that show an average 5-10% increase between old centrino and 2.13 sonoma with 533 memory.
http://www.tomshardware.com/mobile/2...alviso-10.html

So if you take a 10% increase and compare it with other benches, pm should be equal to 3400+, not that much superior.
ahh thanks. I saw that article a while back, but when i searched for it earlier i couldnt find it under the cpu section (who would have thought it was in the mobile section ;p). I looked at most of the benchies, and the performance goes from as small as 2% to 13%.
Reply
Reply
post #72 of 99
4/27/05 at 11:57am
As I see it, the main real ADVANTAGE Turion has over Pentium M is in cost. The biggest deal to me isn't that according to the review, the Pentum M performed better than the Turion, but that the Turion performed comparably (for the most part) and will cost a lot less. Thus, you could get a system with the same specs for hundreds of dollars less with a Turion vs. a Pentium M.
Reply
Reply
post #73 of 99
4/27/05 at 2:14pm
Turion's got SSE3 and 64Bit support too. Wait till next year when 64biy versions of popular programs come out - like maybe 64bit WinRar with measurable performance over 32bit because of the new 64bit instructions.
Reply
Reply
post #74 of 99
4/27/05 at 2:20pm
Thread Starter 
Quote:
Originally Posted by HardBall
I know all that full well, if you have ever read my previous posts you will know that. Such as these:

http://notebookforums.com/showthread.php?t=76011
http://notebookforums.com/showthread...hlight=533+133

I have posted repeated on these boards about the difference between system clock, memory clock, FSB. 2.1 is a common shorthand for 2.13333333333333333 GHz, and everyone in the forum knows that fact. Please don't get so emotional when someone uses a shorthand to indicate the speed of the FSB of your dear old PM, or slighting the CPU clock by one thirtieth of 1GHz.

My point is that none of those other benchmarks posted used Alviso, there is a difference. You should know that. So wait until anandtech posts their Turion benchies and we'll see how it goes.
Reply
Reply
post #75 of 99
4/27/05 at 4:14pm
Quote:
like maybe 64bit WinRar with measurable performance over 32bit because of the new 64bit instructions.
WinRAR is not something I base buying a system on.

Tbithe Pentium M is an amazing little chip that offers performance all out of scale with its power consumption and its technical specs. If Intel were smart, and could overcome its pride, it would drop the P4 netburst deadend and start really improving the P-M.

The P-M is an ideal candidate for multicore simply because you can fit more cores per watt on a die.

The really need to boost its FPU, add SS3, add onchip memory controller, and at somepoint add x86-64
Reply
Reply
post #76 of 99
4/27/05 at 4:37pm
Quote:
Originally Posted by MichaelX30
My point is that none of those other benchmarks posted used Alviso, there is a difference. You should know that. So wait until anandtech posts their Turion benchies and we'll see how it goes.
OK, Michael, I'll give you that; the bickering between the two of us about that is really pointless, and I shouldn't have started it in the first place.

But even given the Sonoma 10% increase in speed; which is quite generous, considering the link above by gilgalasvf, and the fact that FSB is only one of a number of factors determining the speed of the system. For example: there was little speed bump between an 800MHz and 1066MHz Prescott, nor was there much difference between the 1600MHz HTT in the Clawhammer and the 2000MHz HTT in the S939 Newcastle/Winchester.

Even given it had 10% improvement, and given in many past benchmarks, the 3400+ actually scored somewhat higher than the PM 2.1 400MHz, 10% increase would certainly not bring a PM 2.13 533MHz to 130% performance of the 3400+, in the exact same games such as FarCray, HL2, UT2004 and the like. From that fact alone, there seems to be something wrong.

Plus the more obvious fact that the test had a heavily subjective nature to it, and was slanted to test certain aspects of gaming such as scene rendering, to suit the strong aspects of the PM's performance.

Michael, I do get your point, and agree with you that the Sonoma will perform somewhat better; but definitely not on that kind of scale that the review was suggesting.
Reply
Reply
post #77 of 99
4/27/05 at 9:56pm
Quote:
Originally Posted by Djembe_Rob
As I see it, the main real ADVANTAGE Turion has over Pentium M is in cost. The biggest deal to me isn't that according to the review, the Pentum M performed better than the Turion, but that the Turion performed comparably (for the most part) and will cost a lot less. Thus, you could get a system with the same specs for hundreds of dollars less with a Turion vs. a Pentium M.
If the Turion starts to look like a serious threat to Intel's dominance in the mobile space, we'll see any price advantage slim down really quick. Intel has the advantage of being 10 times AMD's size (and that's an understatement); they will gladly take a smaller profit per chip to maintain volume.

This is, of course, good for all of us: a hungry AMD with a solid mobile offering will hold Intel's pricing down there too.

I agree with your point, though: performance at least in the ballpark of the P-M, probably not as good on the power front (but gads better than a P4 or full-blown A64), 64-bit capable today, and all for a bit less than a Centrino offering.

You can't complain about that.

cheers all
Reply
Reply
post #78 of 99
4/27/05 at 10:06pm
Quote:
Originally Posted by Woland
If the Turion starts to look like a serious threat to Intel's dominance in the mobile space, we'll see any price advantage slim down really quick. Intel has the advantage of being 10 times AMD's size (and that's an understatement); they will gladly take a smaller profit per chip to maintain volume.

This is, of course, good for all of us: a hungry AMD with a solid mobile offering will hold Intel's pricing down there too.

I agree with your point, though: performance at least in the ballpark of the P-M, probably not as good on the power front (but gads better than a P4 or full-blown A64), 64-bit capable today, and all for a bit less than a Centrino offering.

You can't complain about that.

cheers all
Actually, it may or it may not. Athlon 64s are very nicely priced in the desktop, and it's for a simple reason - AMD doesn't have the capacity to steal a large portion of Intel's sales. It doesn't help that the majority of people go "There are processors besides Pentiums? I thought those were only in Macs!"
Reply
Reply
post #79 of 99
4/27/05 at 11:47pm
Thread Starter 
Quote:
Originally Posted by HardBall
OK, Michael, I'll give you that; the bickering between the two of us about that is really pointless, and I shouldn't have started it in the first place.

But even given the Sonoma 10% increase in speed; which is quite generous, considering the link above by gilgalasvf, and the fact that FSB is only one of a number of factors determining the speed of the system. For example: there was little speed bump between an 800MHz and 1066MHz Prescott, nor was there much difference between the 1600MHz HTT in the Clawhammer and the 2000MHz HTT in the S939 Newcastle/Winchester.

Even given it had 10% improvement, and given in many past benchmarks, the 3400+ actually scored somewhat higher than the PM 2.1 400MHz, 10% increase would certainly not bring a PM 2.13 533MHz to 130% performance of the 3400+, in the exact same games such as FarCray, HL2, UT2004 and the like. From that fact alone, there seems to be something wrong.

Plus the more obvious fact that the test had a heavily subjective nature to it, and was slanted to test certain aspects of gaming such as scene rendering, to suit the strong aspects of the PM's performance.

Michael, I do get your point, and agree with you that the Sonoma will perform somewhat better; but definitely not on that kind of scale that the review was suggesting.

It also depends what graphic cards were used in the other benchmarks. If the AMD setup used a 9800 and the P-M used a 9700 then there you go. I say we just cool down till we get some more benchies.
Reply
Reply
post #80 of 99
4/28/05 at 12:05am
Cool down forever. You are really bickering about facts that which are known to not be reliable... so why bother? Wait until someone we know as reliable gives a solid run-through of the Turion and compare it to P-Ms. We all have seen A64 performance vs. P-M, so a simple head to head of the latest chips (an O/Ced ML-37 vs an O/Ced 770 would be a good match) with not comparable mobos but mobos that would be sold with the unit (i.e. the SiS). Same goes for GPU/VRAM (so an actual battery mark can be taken... hopefully they'll have x850s!! *whimper*)
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Notebook Forums - General
NotebookForums.com › Forums › General Notebook Discussions › Notebook Forums - General › Have you all seen these Turion Benchmarks?