Features - Lenovo ThinkPad Edge 15: Intel Core i5 vs. AMD Phenom II

[HEXUS]

AMD has revitalised its mobile Vision with the Danube platform, but is it a credible alternative to Intel's Arrandale?

Read more.

[kingpotnoodle]

Wow that's a pretty damning assessment of AMD's offering. Makes you wonder why Lenovo bothered given that these are almost the same price to buy!?

Llano is going to have to be good to beat/overtake Sandy Bridge.

[chis]

Excellent work chaps. More of these "direct" comparisons where possible please.

As for the Edge, Lenovo still need to prove if their "cheaper" non-Thinkpads and lower-end Thinkpads like this have any longevity. I still have a 1GHz P3-based Thinkpad from nigh on 10 years ago that works perfectly well... and a P3-700 TP from even further back that still works! (Runs Lubuntu 10.04 excellently, too!)

And also, no expresscard slot: no chance of using my RME HDSP audio card... sucks!

[Jimbo75]

Wow that's a pretty damning assessment of AMD's offering. Makes you wonder why Lenovo bothered given that these are almost the same price to buy!?

Llano is going to have to be good to beat/overtake Sandy Bridge.

Hardly. The Phenom has 3 cores on 45nm, the 4250 is based on 55nm tech. Compare that to the dual core 32nm intel cpu and 45nm graphics and it's surprising the difference isn't a lot more. 15 mins isn't a lot and neither has great battery life anyway.

The Phenom just needs to be a little bit cheaper, maybe another £50 or so.

[mateau]

Actually the two Lenovo laptops are NOT EQUAL.

The AMD version was using Windows 7 Home Premium 64 and the Intel version was using Windows 7 Professional 64. Win7 Pro 64 has an XP mode was that on or off? Win7 Home 64 does not.

The Phenom is also running 30% slower than the Intel Core i5. Which is very far from damming.

The test is far from valid.

Are the Benchmarks multi-thread benchmarks or single core? I don't know this and the question is not rhetorical but if the benchmark software doesn't excercise more than a single core then why was this comparison made?

So what do we really have? Factoring then for the 30% lead in flops the Intel Core i5 2.2 gz is essentially in the same performance envelope that is exhibited but the AMD Phenom 1.8gz cpu. All things being equal.

Including power usage!!!

So the question really is this?
Is this difference in performance worth $200.00, or 30-35% of the entire value of Lenovo laptop?

Yes the Intel Core i5 is faster 2.2gz > 1.8gz. That's a no brainer.

So what are we trying to say here? That a slower cpu gets beaten by a faster cpu running a better operating system?

[Parm]

mateau, both notebooks were configured and tested with a 64-bit install of Windows 7 Professional.

In regards to the benchmarks, both Geekbench and Cinebench are multi-threaded.

[this_is_gav]

The difference in performance between HP and Pro ought to be almost exactly zero.

Still, value for money wise, the Intel system could be brought down to almost the price of the AMD one if you can configure it with HP.

However, the biggest performance differentiator, the hard drive, wasn't tested here - indeed it wasn't even mentioned beyond the table of specs. The AMD one has a 7,200rpm drive while the Intel a 5,400rpm one. The difference isn't as big in laptop hard drives as it was with desktop drives, but it's very noticeable in real-world use.

[kingpotnoodle]

So what are we trying to say here? That a slower cpu gets beaten by a faster cpu running a better operating system?

What we are saying here is that 2 laptops with very similar prices are showing a great difference in performance.

The AMD one is slower (in some benchmarks by a large margin), has less battery life and weighs more. Especially considering it has the faster hard drive thats a pretty bad result. Performance per watt and especially per $ are worse on the AMD machine.

Doesn't matter that AMD clock speed is lower or that they are using older manufacturing technologies (AMD don't have a ~2.2GHz 32nm CPU available at this price point, this is their best available), this is not a test of clock for clock performance, it's a whole platform test comparing 2 current and competing solutions at the same price point... "if I had £700 what's the best buy?" and the Intel machine wins hands down.

The chassis are otherwise equal, so usability concerns such as screens, keyboards, style etc are irrelevant. AMD can't produce a platform that equals Intel's performance within the same cost and power budgets, therefore their technology is worse and I see little point in Lenovo making the AMD variant, simple really.

It's nothing against AMD, I actually just ordered a Dell M101z - 1.3GHz Athlon II Neo CPU & 4225 graphics, because at the £500 price point I think it to be the best balanced ~1.5Kg machine available now (taking into account it's supposedly robust construction and excellent keyboard which for me are deal makers). Maybe I will post a review at the end of the month when it turns up.

[mateau]

mateau, both notebooks were configured and tested with a 64-bit install of Windows 7 Professional.

In regards to the benchmarks, both Geekbench and Cinebench are multi-threaded.

Your artical is filled with too many errors to be taken seriously.

You originally stated that the AMD Lenovo was running Win 7 Home 64 and the Intel Lenovo was running Win 7 Pro 64.

1. After reading my post and discovering that the 2 Lenovos were running 2 different Operating Systems you decided to cover up that little inconsistancy by editing that problem and declaring that they were the same. You just lied to your readers. Win 7 Pro runs 64 bit multithreaded code a little bit differently that Win 7 Home, or so I've been informed. I could be wrong as I'm not a Win7 expert. This would be a good topic for a future article if you can be trusted not to edit your results to fit your thesis.

2. The AMD Lenovo was running a 7200rpm hd and the Intel Lenovo a 5400rpm hd. The 7200rpm hd burns more energy than does the 5400rpm. Lets not even try to argue seek efficiency. Battery life has always been the issue with fast drives.

3. Pricewatch has the AMD Phenom very cheap compared to the Intel Core i5. The issue is really Lenovo marketing a mid perfomance laptop in the performance niche. You caught them at it, don't blame AMD.

4. Intel is playing a marketing game with Core i5 Turbo Boost. That 2.26 gig cpu is really a 2.53 in disguise, the 2.4 is a 2.93 gig. AMD lowers the clock speed for mobile CPU's Intel "idles" at the advertised and rated speed then increases to the "Turbo Boosted" speed. Basically Intel is sandbagging their performance as a golfer would sandbag his handicap to "play down" in a money tournament. And the media buys it. AMD Penom does not have good energy management. But I would like to see a test

I read your article and the inconsistancies just are too great for it to be completely credible.

So the question remains unanswered.

[Parm]

mateau, the Windows 7 listing was merely a typo. Both systems were tested as shipped with Windows 7 Professional.

Thanks for the feedback nonetheless.

[Zhaoman]

So the question remains unanswered.

What question remains unanswered?

The article compared two different platforms at the same price point and found quite conclusively that the Intel system was better. It was surprising that even the Intel graphics were faster but that's what it shows.

If they kept the CPU clock speed, number of CPU cores, manufacturing process etc etc constant then it would be a comparison between CPU architectures. If that's what you wanted, with everything else kept constant, then there are many articles out there that do this. This isn't one of them and it was never the point of the article.

Imo this was one of the most helpful Hexus articles in while and it was a great concept and very well done.

[Temi_D]

Damn...those results are really ****** up. I thought for sure the AMD system was gonna win all the benchmarks. I mean tri core at 1.8GHz vs. dual core @ 2.26GHz and Intel's HD graphics vs. a 4250. AMD really need to step up their game.

Also Intel's HD graphics just got a whole lot more respect from me

[scaryjim]

@mateau: the i5 runs at 2.26GHz with both cores fully loaded: turbo boost only kicks in when one core isn't utilised. The CPU *idles* at ~ 1.3GHz under EIST, in the same way AMD processors idle at 800MHz under CnQ. Hexus reviewed two equally priced laptops from the same manufacturer - I don't see how you think it's an unfair review.

@Temi: the i5 processor has HT enabled, so can run 4 threads, albeit with a performance hit compared to an equally clocked native quad-core. When Hexus reviewed the Core i5 661 (3.3GHz), it was roughly performance-equivalent to a 2.5GHz Phenom II X4 in multithreaded tests. Based on that, it'd take a mobile Phenom II X4 at ~ 1.7GHz to match a 2.26GHz mobile i5 (assuming the relative performance is the same as desktops). So it's not surprising that a 1.8GHz x3 returns about 75% of its performance - it's exactly what you'd expect based on the comparitive performance of the desktop equivalents!

[Jimbo75]

The only issue I have with the review is the use of 3dmark 06 which is well known to be more cpu-bound than gpu.

The 4250's graphics are better than the i5's and that would show in most games, and probably 3dmark vantage too, but there isn't a lot in it these days.

[kingpotnoodle]

The only issue I have with the review is the use of 3dmark 06 which is well known to be more cpu-bound than gpu.

The 4250's graphics are better than the i5's and that would show in most games, and probably 3dmark vantage too, but there isn't a lot in it these days.

That is a good point actually, they should have used a real world game benchmark.

[semo]

why did they have to make include only one 3.5mm jack. It's not like they are short on space on a 15" body. Can't use headsets with a mic/microphone combo jack