USB as Power Boost? Vista Tips Discuss USB as Power Boost? in the Windows Vista forums; Yes, ratings are not everything.. I usually look for the "Vista Certified Sticker".. Not always 100% that way, (could be Fake) but I do feel better when I ...

[Snuffy]

Yes, ratings are not everything.. I usually look for the "Vista Certified Sticker"..
Not always 100% that way, (could be Fake) but I do feel better when I see it... A-data PD7 has the sticker... list on the TW site..
the site which shipped the PD9 said it also was Certified..only it does not list PD9 on the manufacture site... I think that is a USA number..(LMAO), who knows except it does pass every test.. and the speeds are very close ea time..it is tested...

[Znod]

Originally Posted by Znod

Quoting myself: "Here are the key related requirements given by Tom Archer again:

The device must be able to do:

3.5 MB/s for 4 KB random reads uniformly across the entire device and

2.5 MB/s for 512 KB random writes uniformly across the device.

In this regard, we can get some data comparable to to these, but not always exactly comparable. I'll elaborate later. Have to go to the gym, etc., now."

Nevermind, I lost the mind part of nevermind temporarily. I'll think more about what I am thinking.

[Snuffy]

For each test a file is created in the root directory of the selected disk or in the current user’s application directory (as specified). When running PerformanceTest on Windows Vista (without elevated administrator privileges) it is recommended that the User directory be selected (as generally the user will not have write permission to the root directory of the drives).

So to run really it should be in non admin login?

[Znod]

Originally Posted by Snuffy

For each test a file is created in the root directory of the selected disk or in the current user’s application directory (as specified). When running PerformanceTest on Windows Vista (without elevated administrator privileges) it is recommended that the User directory be selected (as generally the user will not have write permission to the root directory of the drives).

So to run really it should be in non admin login?

Well, PT works fine for me. Since it works fine for me, I never worried about what directory was being used.

[Znod]

Originally Posted by Snuffy

Yes, ratings are not everything.. I usually look for the "Vista Certified Sticker"..
Not always 100% that way, (could be Fake) but I do feel better when I see it... A-data PD7 has the sticker... list on the TW site..
the site which shipped the PD9 said it also was Certified..only it does not list PD9 on the manufacture site... I think that is a USA number..(LMAO), who knows except it does pass every test.. and the speeds are very close ea time..it is tested...

Yep, my Apacer came with the sticker.

[Znod]

Originally Posted by Znod

Next Installment:

Since ReadyBoost is supposed to shine on disk operations, I decided to focus only on the 6.1 disk tests for my second installment. I followed essentially the same procedure as above.

Results:

Averages of best 5 of 7 with ReadyBoost:

Partial Composite = 407.0 (Best = 421.4)
PassMark Rating = 81.42 (Best = 84.3)
Sequential Read = 54.16 (Best = 54.4)
Sequential Write = 50.58 (Best = 54.3)
Random Seek + RW = 7.78 (Best = 7.8)

Results: Average of best 5 of 7 without ReadyBoost

Partial Composite = 370.8 (Best = 401.1)
PassMark Rating = 74.16 (Best = 80.2)
Sequential Read = 53.9 (Best = 54.0)
Sequential Write = 45.68 (Best = 54.1)
Random Seek + RW = 2.93 (Best = 2.96)

To date, these tests show ReadyBoost in its best light. It appears to beat my non-ReadyBoost setup most significantly on sequential wirtes and random seeks plus read/write. An interesting finding is that the PassMark Partial Composite score for ReadyBoost is, on average, 36.2 (407.0 - 370.8) points higher than the corresponding non-ReadyBoost score. Noting, from my earlier tests, that the difference in the overall PassMark Composite score with and without ReadyBoost is only 4.28 (687.5 - 683.22) points raises an important issue. Does ReadyBoost hurt performance in other areas? Investigating this issue is my next task.

Here is my 3rd installment--following up on the question I raised at the end of the above post. Incidentally, the answer to the question is no, and I should have known so when I raised the question. That is, above when I calculated 36.2 (407.0 - 370.8), the calculation should have been: 7.24 = [.2(407.0 - 370.8)] since the PassMark weight for its disk test composite is 20% (i.e., for example, .2(407.0) = approx. 81.42--see above).

But, anyway, below are the results of my latest test--a full test. My results show that, indeed, a ReadyBoost Device excels, as we would expect, on disk tests and has little effect otherwise. And, as should be expected, my results show that ReadyBoost does not appear to hurt performance in anyway (i.e., any apparent "hurts" should turn out to be statistically insignificant if enough tests were done to draw statistically valid conclusions). Still, a few questions are raised when looking at the 3D graphics results.

Note that my Apacer flashes when in use; thus, by looking at the device, one is somewhat able to infer which tests are likely to make differences in the PassMark overall score and which are not. But, in this regard, the Apacer appears to make a difference on the memory tests, although no flashing occurred during these tests. Hmmmm????

As before, each average given below is the average of the best five of seven passes. And, in general, my test procedures were as described in my earlier test-related posts (two more above). Note that the PassMark test weights are: Disk (20%), CD/DVD (9%), Memory (18%), 3D Graphics (12%), 2D Graphics (14%), and CPU (27%). Thus, the difference in the composite-score averages of 5.96 (688.08 - 682.12) can be calculated as: .27(.14) + .14(.14) + .12(-1.12) + .18(5.94) + .2(24.56) + .09(.04) = 5.91 (i.e., approximately 5.96)

[Znod]

Originally Posted by Znod

Here is a good read on flash drives and ReadyBoost:

http://blogs.zdnet.com/Bott/?p=186&tag=nl.e622

QUED on the Apacer Handy Steno. My new 4 GB should arrive by Fed Ex today. Hope this one doesn't break like my first one did. I'll be posting my test results here.

Did anyone notice the Bott-claimed requirements for a flash drive given below and in the link above. They are different from the one's given by Archer earlier. I wonder which are correct. Bott's blog is more recent, but who knows. Here is what Bott says:

"To be used as a ReadyBoost device, your flash drive has to pass several tests, including available free space, write performance, and random read performance. When you connect a supported flash device to your system and choose the Speed Up My System option, Windows Vista runs a quick performance test to see if the device meets minimum standards required for ReadyBoost. Those standards are:

* 2.5 MB/sec throughout for 4 KB random reads
* 1.75 MB/sec throughout for 512 KB random writes

These results must be consistent across the entire device. In addition, the device must be at least 235 MB in size (although you can designate less than the full space on the drive for the ReadyBoost cache).

If any of these tests fail, the drive is rejected."

[felix8406]

You can get that info from the horse's mouth by downloading a .doc file from http://www.microsoft.com/whdc/system...perfaccel.mspx. It's pack with some excellent information about ReadyBoost, SuperFetch and ReadyDrive.

Here are two quotes: "Flash storage devices must meet minimum performance requirements to support ReadyBoost, including 2.5 MB/s throughput for random 4K reads and 1.75 MB/s for random 512 K writes."

"Higher performance requirements must be met for the device to be designated as “enhanced for ReadyBoost”: 5 MB/s for random 4K reads and 3 MB/s for randmon 512 K writes. Microsoft recommends that users use devices enhanced for ReadyBoost to take full advantage of the benefits provided by ReadyBoost."

[Znod]

Thanks for the additional info felix. Your supplement helps.

Given that I am using an enhanced device, what do you think about the minimal overall performance improvement provided by ReadyBoost shown in my last set of test results. Of course, ReadyBoost helps when it comes to disk operations.

PassMark gives disk operations a 20% weighting in its composite score formula--which might imply that the company thinks that about 20% of one's computing time is spent on disk operations. If so, then the performance increase I show for disk operations seems good.

However, I can never perceive any performance differences for disk operations when I run the PassMark disk tests with and without ReadyBoost. So, I am persuaded that on a highly capable machine one will never notice the effects of ReadyBoost.

I am going to leave ReadyBoost testing on laptops to you guys. In this regard, I have never been able to perceive a performance difference in my disk tests on my laptop with and without ReadyBoost. While interesting, I think ReadyBoost is more like ReadyBust even though performance differences can be found with test programs. Let me know if you can perceive any performance differences when running the disk-operation tests (or the memory tests) when you do your tests.

[iMatt]

One question: What about USB 2.0 microdrives? I have one of those Philips USB 6Gb microdives. Will this work with ready boost or will readybosst need to work exclusively with flash based storage?