Windows 7 review - notebookcheck

I have to admit, one of the pleasures of reading reviews at notebookcheck.net - for me - is the always idiosyncratic German to English translation. And this one doesn't disappoint.

http://www.notebookcheck.net/Practical-Review-Windows-7-RC1.18274.0.html

Not a complaint, btw. Those folks do a lot of good work, and I'm just glad they translate at all. Because mein Deutsch ist nicht so gut.

 

Wait a minute, Windows 7 is not even finished, how can anyone even remotely think of writing a review? It will be obsolete the moment it becomes important: when Windows 7 is out and people have to decide whether to upgrade or not. Well, as a preliminary thing it may still be helpful for some people for the RC already has a pretty high quality.

Without having read that RC1 review (thus not judging it), I offer to explain whatever is incomprehensible. Let's hope I'll be able to produce more idiosyncratic English for your enjoyment.
(no, I did NOT do that noteboocheck.net translation)

Edit: http://www.youtube.com/watch?v=LMQjRkoGH3Y

 

they could call it a preview.

previews are relevant in that they help people formulate ideas about something before they are suddenly faced with a decision.

but usually these shouldn't amount to much more than "looks good so far- check back" or "looks horrible so far- check back"

 

I don't know about that. Let me say first that I know nothing about "those folks", but that review is complete garbage. When I read their random fantasies regarding the stability of various MS OS, I can't help but wonder... Vista unstable? You got to be kidding me! People can complain about all sorts of things with Vista, but instability is not one of them. "Annoyance factor"? What the heck is that supposed to mean? Again, they have some incoherent ramblings about this, but nothing that is of any value to anybody.

 

Your german is better than mine.
But I think it's a little too early to do a "review" yet, it's not even finished!

 

@Pirx: Read their notebook reviews. Specifically their notebook display tests. Who else is that thorough?

That doesn't mean their "review" of W7 is beyond criticism, of course. But one arguably silly OS "review" hardly outweighs all their informative notebook reviews, or their work in benchmarking & cataloging mobile GPUs, etc.

 

I'm sure it will be quite different when it launches, but I'm glad to see that even the RC gives a good impression. I'll be looking forward to its launch!

 

Pirx doesn't mention anything about any other reviews, unless you saw something that I missed. All I saw was a very critical review of their "review".

 

The result of their tests goes along somewhat with my observations: although Windows XP is arguably the more efficient of the three, Vista, and then ultimately (and more successfully) Windows 7, employ superior design features in several areas.

And yes, Vista and W7 are not necessarily less powerful than XP, if you were to compare them on a recent machine. However, I still like to note that XP, or especially Linux distros, offer comparable performance and functionality on a newer machine, and also run quite satisfactorily on older hardware.

All in all, it all comes down to I'm glad W7 is out-performing Vista.

@Pirx As masterchef said, it should be regarded as a *preview*.

 

I have to say that while the hardware reviews can be decent, elements of this review are questionable.

They could have easily benchmarked the installation with all x64 systems, or all x86 systems. As people have pointed out, Windows Vista and Windows 7 are available in 32-bit versions. I find the installation ratings dubious here because the XP system should have a lower score. It installed less data but took more time, looked bad, AND required drivers to be added post-install. I also wonder what constitutes a perfect score here. Seventeen minutes to copy a DVD image to the partition, decompress, and install an OS that works fine out of the box garners an 80% score. The concern here is whether the hardware is the limiting factor in this case.

The Use of Resources section is disgraceful, and demonstrates a mid-90's understanding of memory usage and allocation. The total RAM taken by a modern OS is dynamic, partially based on how much RAM you have available, and is at best a loose indication of system performance. While XP's poor memory management model trained people to bench free RAM, Vista and Windows 7 will prefetch data from the disk into volatile memory so that programs launch faster. The upshot here is that when prefetched memory is requested by another program, the amount of time necessary to release that memory is essentially zero. The amount of memory occupied by prefetched data is proportional to the amount of RAM installed, thus, if they benched with an Octogig machine, they'd see even higher RAM usage in both Vista and Windows 7. Conversely, if they benched with a 1 gig system, they'd see lower memory usage. I'm surprised that NBC would reward points based on how inefficently an OS uses its hardware resources.

In the performance benchmarks, I find it a little odd that while all three operating systems have similar average FPS during gaming, the heavily criticized Vista, and pre-release Windows 7 have higher absolute minimum FPS than XP.

I admit, I don't understand this sentence.

Yes, Windows Vista is the slowest, by a whopping four points (0.032 %)!

As for system stability, I really have to wonder whether these guys understand 64-bit systems, as with their apparent (failed) attempt to install a 32-bit printer driver, as well as running some archaic version (is that version 3?) of Photoshop. Backwards compatibility is great, and I'm all for it, but I am forced to wonder whether they've intentionally generated crashes to lower the reliability score in the screenshot.
Edit: The icon belongs to Photoshop 4, 5, or 6.

On the Compatibility end, I've noticed that many, many reviews will focus on backwards compatibility, but few care about what you get by moving forwards. One good example is Protected Mode in IE7 and IE8. Vista and Windows 7 are also more likely to offer up-to-date drivers for more modern peripherals, especially on x64.

On the subject of annoyance factor, UAC is annoying when you're installing programs. It plateaus in the first day or two when you're setting up your computer. After that, you should notice a decrease - most people don't constantly create or delete folders in Program Files, nor do they need to manually shove some weird driver32.sys into %Windir%. Ever clean up a malware infestation? If UAC can help users catch drive-by installations before they happen, then it makes up for a prompt every time you want to install a program in OS Land.

Vista's Updates take a deplorably long time to run. It is interesting that Vista and Windows 7 have moved away from an IE-centric update scheme, and instead have integrated Updates with an Explorer pane.

 

Hey, the first thing I thought when I had a quick look at that review was: "Well, that's not very thorough...". It seems to mainly consist of irrelevant information to fill up all the gaps between the interesting bits. Much of all the information is just not researched thoroughly enough. And some of it is just plain wrong.
You're right, their notebook reviews are pretty thorough. However, that Windows 7 RC1 review makes me wonder if they really know what they are doing.

Well, on the other hand I'm used to reading a magazine called "c't" whenever I'm looking for reviews, and they have a very good reputation, are very thorough and know what they're talking about: http://www.h-online.com/security/Cracking-budget-encryption--/features/112548

Hm, I wonder if http://www.h-online.com/news/ is of interest for you, chris-m.

 

I tried Windows 7. My impression was that some GUI solutions were borrowed from Linux world. Is it true?

 

The Linux world, Windows and OS X are all borrowing from each other. That's not a bad thing.

 

Really? I am not aware of it. But then OS borrows from each others idea now and then. Can you link us to the rumor?

cheers ...

 

prefetch isn't the perfect model of memory management.

1. keeping spare memory filled with data that you may not use can cause hard drive thrashing (annoying, can waste hdd i/o resources)

2. if you have a lot of memory available and windows starts prefetching, and then you have a memory intensive program run (intensive of bandwidth, not space) now you have an issue

---

sure, there are also many upsides to prefetch. its not like prefetch is "mostly bad", but it certainly isn't the perfect model of memory management, either.

 

it isn't a rumor, this is a fact. they don't borrow code flat out (except sometimes from open source code made publicly available for free) but they do borrow ideas and design.

 

Do you have a link to an article that discusses Microsoft borrowing from Linux?

 

You don't need link or rumor. All you need is to have your experience with
installing Windows 7 and different Linux distros. Then you will see.

 

The sad thing is: Vista is meant to only Superfetch contents to RAM if the HDD is idle enough. However, it seems like this just doesn't work! I have often found a Vista installation slowing down the user because it was working on the harddisk all the time, with disabling Superfetch curiously ending in speeding up the computer. Seems like Microsoft didn't get the right threshold.

Be aware, "Prefetch" and "Superfetch" are two different things. Superfetch is new to Vista, whereas Prefetch (kind of a dumb Superfetch) has been there since XP.

Superfetch will immediately free RAM that's in use by it whenever any other application is demanding more RAM.

I don't see why bandwidth should be an issue, with the disk being the limiting factor. What exactly do you mean?

 

All GUIs borrow from each other, whether they try to or not.

 

I installed W7 and many Linux distros and many OS X. Sure one can feel that OS developers (products) "borrowing" each other ideas and one can discuss and talk within tech circles. But to saying it out right - in writing - I would like to read it black and white.

Link me up

cheers ...

 

I find it ironic that Linux ended up getting the wobbly windows that were first demoed with Longhorn.
http://www.extremetech.com/article2/0,1697,1154262,00.asp

 

But which came first?

 

The earliest implementations that I could find on the Linux side are from 2005. XGL came out a year after that.

 

I think the true nature of this "review" is exposed with the screenshot that they took of the Performance and Reliability monitor.
http://www.notebookcheck.net/typo3temp/pics/15d9cf330c.jpg

Installing 32-bit drivers on a 64-bit system? Yeah, that'll cause an error. Anyone looked up what version of Photoshop they failed to run there? The absolute latest possible version is 6, released back when Windows 98 was the dominant consumer OS, and Win2K and WinME were the bleeding edge. The earliest version with that icon came out in 1996. Yup, 13 years ago. If that's what it takes to produce errors in Windows 7, then this review is pathetic.

 

Agreed, Relativity.

The instant I hear somebody bring up Vista's "Wasteful Memory Usage", I throw their review out the window.

 

which code did they borrow/steal?

can you name more than one and show an example?

can you show the link and proof of said so called fact

 

People are not saying that code is stolen. I don't get why people are up in arms about this statement that all OS's borrow from each other. It's simple, when you have competing ideas, competitors will borrow ideas from each other, either consciously or subconsciously and mold it to fit their vision, that is all they are saying.