Intel's Pentium M 760 versus AMD's Turion 64 ML-44

[rettihSlluB]

Looks like I'm still alive (let's see till when):

http://techreport.com/reviews/2006q1/pentiumm-vs-turion64/index.x?pg=1

The Turion 64's on-die memory controller is hampered by the single channel DDR400 limitation, while the Pentium M's high memory bandwidth is handicapped by its slow front-side bus.

It's really impressive to know that Turion 64, with half the cache and running on single channel, gave the Pentium M a tough fight (and in some cases beating it). It's hard to believe that dual core Yonah would be competitive against a dual core Turion 64, which will feature dual channel and DDR2.

From a power consumption perspective, the Turion 64 surprised me. Yes, our Turion 64 test system consumed a third again more power than the Pentium M system at 100% CPU load, but unless you're using your laptop to crunch that F@H work unit on the plane, maximum power consumption isn't usually all that important. For typical use, it seems likely the Turion 64 would be competitive with the Pentium M on the battery life front, as well.

This proves that Turion 64 is cooler than Pentium M (when running iddle).
If they ever used the MT instead of the ML, power consumption might be even lower than that of Pentioum M.

 

[CompGeek]

Find a review from another site. Those guys are AMD bias. Just look at the 955 EE review (955 VS FX 60(but x2 4800 is mentioned)) there then look at Xbit/THD/AT review.

They carefully chose progs or versions that don't work so good at intel. I call that BS if you ask me. Xbit, Anand Tech and Toms Hardware have come up with fair reviews these guys from "Tech Report" are AMD fanboys. No need to post links that lead to such a biased site. Compare Intel 955 EE with the X2 4800 at encoding. In any of the trustfull sites Intel wins at least 1 benchie while in TR it doesn't.

There is nothing special about the Turion,it consumes more energy and now there is a Core Duo that kicks its arse.

 

[luminaris]

While I do agree with the first part about the pentium having a slow frontside bus, I don't agree with the power consumption part. IMO, no laptop hardly ever sits at idle speed. Yes, they are designed to be very efficient which means throttling back the power but as soon as you execute anything, that changes. With all the bloatware and junk that is shipped with every new laptop running, your CPU has to work while this junk is running in the background.

 

[Ycon]

Turion is so poor, it can never keep up with a Pentium M. But that site basically tells you that Turion is better than a Pentium M which indicates that the site is cr*p.
Another thing: Why is it the top-of-the-line Turion against the 3rd fastest Pentium M? Someone trying to put AMD into a better light? They also try to find excuses (which btw are a laugh) for every loss of the Turion while they comment every Pentium M "loss" with "Turion crushes" "Turion roars back" *lol* AMD Fanboys...

 

[rettihSlluB]

Find a review from another site. Those guys are AMD bias. Just look at the 955 EE review (955 VS FX 60(but x2 4800 is mentioned)) there then look at Xbit/THD/AT review.

They carefully chose progs or versions that don't work so good at intel. I call that BS if you ask me. Xbit, Anand Tech and Toms Hardware have come up with fair reviews these guys from "Tech Report" are AMD fanboys. No need to post links that lead to such a biased site. Compare Intel 955 EE with the X2 4800 at encoding. In any of the trustfull sites Intel wins at least 1 benchie while in TR it doesn't.

There is nothing special about the Turion,it consumes more energy and now there is a Core Duo that kicks its arse.

Turion is so poor, it can never keep up with a Pentium M. But that site basically tells you that Turion is better than a Pentium M which indicates that the site is cr*p.
Another thing: Why is it the top-of-the-line Turion against the 3rd fastest Pentium M? Someone trying to put AMD into a better light? They also try to find excuses (which btw are a laugh) for every loss of the Turion while they comment every Pentium M "loss" with "Turion crushes" "Turion roars back" *lol* AMD Fanboys...

I know that truth hurts, but you have to accept it when you see it. 8)

While I do agree with the first part about the pentium having a slow frontside bus, I don't agree with the power consumption part. IMO, no laptop hardly ever sits at idle speed. Yes, they are designed to be very efficient which means throttling back the power but as soon as you execute anything, that changes. With all the bloatware and junk that is shipped with every new laptop running, your CPU has to work while this junk is running in the background.

Well, it looks like you're the only reasonably one to argue here.
It might be true that most of the time you don't keep your laptop iddle, but remember that the article was reviewing a Turion 64 ML model which draws more power than the MT derivative. I'm pretty sure if they've ever used the MT, things would have look a lot embarrassing for the Pentium.
And if your question is why they didn't do that from start, it's because it was a performance test.

 

[luminaris]

I still don't believe you man cause your such a bullshitter. lol

I think it is a step in the right direction for AMD in the mobile platform. Every article I've read so far indicates that the Turion has potential but like the Pentium M, it has its flaws too. I think if anything and the way Intel has been with their recent P4's, AMD will get these flaws worked out and prevail.

 

[Levium]

Turion is so poor, it can never keep up with a Pentium M. But that site basically tells you that Turion is better than a Pentium M which indicates that the site is cr*p.
Another thing: Why is it the top-of-the-line Turion against the 3rd fastest Pentium M? Someone trying to put AMD into a better light? They also try to find excuses (which btw are a laugh) for every loss of the Turion while they comment every Pentium M "loss" with "Turion crushes" "Turion roars back" *lol* AMD Fanboys...

http://www.tomshardware.com/2006/01/16/will_core_duo_notebooks_trade_battery_life_for_quicker_response/

Lets think about the point of owning a laptop. Portability. You don't want your laptop sucking dry in less than an hour, that'd be completely pointless. So in all honesty, the turion is the better choice for battery life. I had a pentium m laptop, a dell 5150, and it was a piece of crap. Could barley run microsoft office without sucking the battery dry in less than an hour. Now I'm running an Acer TravelMate with a turion in it, and it goes around 3 hours no problem. I'm not a fanboy of either, but personally I would choose an amd notebook over an intel one due to the pure fact of actually having "mobility". And btw, when you people start bashing a websites credability, it puts your own on the line. Insults are a pathetic last stand. And also, check the new york stock exchange and see which companies stocks are worth more, imo that'd be the company to invest in...

 

[CompGeek]

Why don't you look for yourself. All important sites have Intels 955 chip win over the 4800 in decent amount of benchies while in this one none except the obvious SySandra.
I don't need to be a fanboy to see that there is something not right with these comparisons. A site that favours a certain brand is not a credible site.
If everyone but X says Y and X says Z who should we believe? Well we should believe the world.
One review is usualy enough to prove a site's worth especially when it's obviuosly forcing the results.

The only problem with Core Duo is a leakage that shall be fixed (if it hasn't already, there is a fix by microsoft) . Intel laptops are known for better performance per watt and no recent turion changed that .
http://www.tomshardware.com/2006/01/16/will_core_duo_notebooks_trade_battery_life_for_quicker_response/page22.html
Were is your mobility gone?

 

[luminaris]

I do find that most sites are biased in one way or another. I for one, do not rely on benchmarks or testing from these sites as, I find it rather confusing to go to one site and see an article praising one product and yet, go to another one and they are praising the other.

To me, the problem is this, if one person who uses a particular product for their own use, tends to not give any credit to the other especially while performing reviews. If I were to bring an Intel machine into an AMD users house, chances are, they've already formed an opinion about that Intel machine just by what they've read or may have experienced in the past. To me, that's just plain biased. Its present all over these forums and it always will be. I don't expect AMD users to jump on the Intel bandwagon when Intel starts kicking AMD around and it will happen sooner or later. It always does.

To me, reviews are biased no matter what so that's why I don't rely on them. What I rely on is what's on my bench in front of me. 4 machines, 2 Intel and 2 AMDs that I can compare.

 

[CompGeek]

Good point.
You can't really have a clear view on 2 products unless you own them and use them. And that way you'll create a personal opinion that may differ from the rest.
Though benchmark should give a foggy view on the subject.
For example if X>Y by a long shot in all benchies , X is better than Y .

 

[ltcommander_data]

It's been a while since I saw a Dothan vs Turion comparison. The problem I had with this comparison was that it's conclusion was fundamentally flaud. They claim that their results are representative of a labtop, but their comparison isn't set-up as such. A labtop's battery life is dependant on a lot more than just the processor. Motherboard and graphics cards are important power users. The advantage that Intel has with the Centrino platform is that they can design a lot of the system with power savings in mind. Component choice makes a lot of difference when talking about power consumption so it's difficult to compare. However, with most Intel Dothan labtops being Centrino certified, it's more likely that you will receive good battery life.

In the review they used AOpen i915Ga-HFS motherboard. If they had really wanted to be more reliable toward how much power a Centrino platform would use then they should have used the AOpen i915GMm-HFS. The results would then be more applicable toward labtops. The best of course would be a direct labtop to labtop ccomparison but cross platform standardization would be very difficult. I will admit that the results are reflective of desktop use, but not mobile use.

What I found interesting in terms of architecture was the fact that the 760 acquits itself very well against the ML-44 especially considering the Turion was operating 400MHz faster. Dothan generally appears superior in performance per clock although it's hampered by it's weak floating point performance. The 760 also has the advantage of being $60 cheaper.

 

[dvdpiddy]

Comp geek is biasd towards intel!

 

[luminaris]

Those are good points man. The thing is too and as you mentioned, there are other factors involved. One of the things I had happen when I bought my one laptop was the battery would drain so fast even when fully charged. Turned out, the battery was bad. I had it replaced and it worked good.

Conditions of the battery, motherboard, environment of the machine, temperatures etc. are all factors in determining how the machine is going to run. Efficiency or none, all those things are factors.

 

[WINDSHEAR]

Comp geek is biasd towards intel!

he is not

and

Dvdpiddy is biased towards AMD! 8O OMG!!!

 

[dvdpiddy]

WELL YOUT BIASED TOWARDS INTEL TOO!

 

[Levium]

Comp geek is biasd towards intel!

You're ate up dvdpiddy...Being biased towards amd or intel is like being biased towards peanutbutter or jelly. Each has their advantages and disadvantages that you can't ignore -.- It's pointless to dive straight into either platform and stick with it soleley, because things will change. Maybe intel will release some uber ass kickin processor that amd can't keep up with, 2 years later amd releases something the world's never seen and intel goes bankrupt, who knows...Just gotta get what ya need...

 

[dvdpiddy]

WHY DO YOU THINK MORE INSTRUCTIONS PER CLOCKCYCLE AND BETTER RAM!

 

[dvdpiddy]

damn it levium your right oh and btw

AMD FANBOY'S UNITE!!!!!

 

[WINDSHEAR]

WHY DO YOU THINK MORE INSTRUCTIONS PER CLOCKCYCLE AND BETTER RAM!

but the Pentium M has precisely that! AND better RAM.... that's why the Pentium M pwns AMD... soon Conroe will beat all existing AMD's....

 

[dvdpiddy]

WHY DO YOU THINK MORE INSTRUCTIONS PER CLOCKCYCLE AND BETTER RAM!

but the Pentium M has precisely that! AND better RAM.... that's why the Pentium M pwns AMD... soon Conroe will beat all existing AMD's.... my New York ass it will amd pwns intel on the desktop market and i ad mit that amd getz pwned on the mobile market.

 

[CompGeek]

WELL YOUT BIASED TOWARDS INTEL TOO!

If Intel's mobile CPU is clearly better than AMDs why not to like Intel?
For its desktop CPUs? Well that's off topic my friend!
You aren't biased if you have reasons to sustain the better product.

 

[Ycon]

Erm... cant we somehow get rid of dvdpiddy? :lol:

 

[luminaris]

Yeah, just give him an original pentium 33Mhz chip and tell him he can't come back to the forums until he gets it overclocked as high as my machine

 

[linux_0]

WHY DO YOU THINK MORE INSTRUCTIONS PER CLOCKCYCLE AND BETTER RAM!

but the Pentium M has precisely that! AND better RAM.... that's why the Pentium M pwns AMD... soon Conroe will beat all existing AMD's....


I disagree.

In 64bit mode the Turion is 20-40% faster which makes it faster than any Single Core mobile Intel CPU.

 

[luminaris]

Ya know, I gotta tell ya, I don't think Intel will ever get the whole 64 bit thing down like AMD has it. Here's a scenario for ya and you'll probably think i'm crazy for saying this but:

1. AMD came out with 64 bit in 2003 I believe.
2. Intel has never grabbed a handle on it
3. Nobody is really making 64 bit applications
4. Intel is now using EMT64 on basically everything now
5. Now i'm really starting to see more and more 64 bit applications
6. Microsoft Vista to join forces with Intel Viiv technology


See a pattern here? I wonder if Intel has some influence on the 64 bit applications and why they're not being developed. I for one would most certainly think we would have migrated to a 64 bit platform in everything we do now. Am I crazy for saying this?