SOLVED ReadyBoost Testing

[BearPup]

Good Day People,
I have used ReadyBoost on my Win 7 x64 Home Premium systems for years; I use it on my 8 GB desktop and 2 GB laptop PCs. But I've never had a way to test the effectiveness of using the ReadyBoost technology, and I'm wondering what exists to test this technology.

Basically, I keep 2, 16 GB flash drives plugged into my quad core AMD rig with 8 GB of RAM, and a 4 GB flash drive plugged into my 2GB RAM Core Duo Intel laptop.

In effect, am I wasting my money on flash drives that I could put to better use (like transferring files)? Are these machines "faster" or in some other empirical way "better" than machines without these ReadyBoost devices?

Any suggestions for testing, or an article on the effectiveness of ReadyBoost, would be greatly appreciated.

Thanks in advance,
BearPup

 

[clifford_cooley]

In my opinion it boils down to this, which is based on fact.

You are better off with true memory or paging to HDD/SSD if you need extra memory. USB flash drives are slow and would not render any benefit over that of faster hardware. ReadyBoost is basically paging to a USB flash drive, which would be slower than normal paging. To be honest, I can't believe ReadyBoost became a thing in the first place.

Edit:
Let me clarify, I have 8GB memory with no Paging enabled of any kind.

 

[BearPup]

What you're saying makes sense for SSD, and Microsoft says so on their web site, "If your computer has a hard disk that uses solid-state drive (SSD) technology, you might not see an option to speed up your computer with ReadyBoost when you plug in a USB flash drive or flash memory card. You may instead receive the message, "Readyboost is not enabled on this computer because the system disk is fast enough that ReadyBoost is unlikely to provide any additional benefit." This is because some SSD drives are so fast they're unlikely to benefit from ReadyBoost.".

By implication though, the same isn't true for standard hard drives. As far as not using a paging file, I have a 1 GB RAMDisk paging file on my 8 GB rig and its in use about half the time according to my HardInfo 7 monitoring software.

So, I guess I need to disagree with you on that point. I'm not sure what to say about the first point though, its kind of why I wanted to do my own testing though, to see how it works on my specific computers. Thanks anyway though.

 

[Digerati]

In effect, am I wasting my money on flash drives that I could put to better use (like transferring files)? Are these machines "faster" or in some other empirical way "better" than machines without these ReadyBoost devices?

IMO, with 8Gb of system RAM, yes you are wasting your money on flash drives.

Are these machines faster? The question is, can you notice any performance improvement? Again, with 8Gb, I suspect not - other than via the placebo affect. IMO, my perception of performance is much more important to me than the results of some artificial benchmark program.

Are the machines "better" with or without? No. I mean, what is "better"? Having more parts that can fail does not improve reliability, or the integrity of your data.

As for paging, I would advise against disabling paging. It is not likely any of us have more expertise in this area than the Microsoft engineers and their super computers they use to determine the algorithms used to efficiently use all memory resources - including the PF. The idea that disabling the PF some how increases performance (by forcing Windows to stuff everything in RAM) is unfounded. Just because you have gobs of RAM does not mean Windows cannot, or will not use the PF efficiently. Therefore, I do recommend you enable the Page File and for later versions of Windows, let Windows manage it. All you have to lose is a little disk space and if that really matters, you need a bigger disk.

That said, moving the PF off the boot drive (if you have more than one physical drive) can improve performance - at least in theory - as Windows can use multiple drives (but not partitions) simultaneously. Leaving a small PF on the boot drive can be beneficial if you need to analyze dump files. Otherwise, a PF on the boot drive is not essential if you have the PF located elsewhere.

Will a computer with lots of RAM work without a PF? Sure. But lots of system memory DOES NOT IN ANY WAY suggest the system will run better without the PF. Evidence suggests otherwise.

Understanding the Windows Pagefile and Why You Shouldn't Disable It is a good article.

 

[clifford_cooley]

As for paging, I would advise against disabling paging. It is not likely any of us have more expertise in this area than the Microsoft engineers and their super computers they use to determine the algorithms used to efficiently use all memory resources - including the PF.

You are leaving out the fact that Microsoft is catering to everyone with a default configuration of which not everyone needs. I've been operating without a pagefile for over 6 months now, and its not the first time I have disabled paging without issues. Issues will only arise if and only when memory usage is maxed out. Then and only then will I enable paging once more. So no thank you, I will pass on reading your link.

 

[Digerati]

So no thank you, I will pass on reading your link.

Okay. That's your choice but I prefer to keep an open mind and be willing to accept that what once was true (or what I once thought was true) may no longer be true.

You are leaving out the fact that Microsoft is catering to everyone with a default configuration of which not everyone needs.

No they aren't. If that were so, you could not move or customize the PF. And the very fact later versions of Windows adjusts the PF in real-time as needed proves that is not true.

No two computers are alike - at least not after they are fired up the first time. For this reason, Windows is constantly adjusting the PF as needed - if you let it.

Issues will only arise if and only when memory usage is maxed out.

Ummm, no. Sorry but that is incorrect. Some programs require the PF. And even with lots of RAM, mapping issues can occur with no PF.

If you can provide a link that shows disabling the PF when you have lots of RAM is a good idea, I will be more than willing to read it.

Just because you have not seen any issues, does that mean your system is better off without it?

Please show us, what are the advantages of disabling the PF?

Does Mark Russinovich - arguably the most qualified Windows expert on the planet - suggest you disable the PF? No.

 

[BearPup]

Ah people...this thread is about ReadyBoost, not the paging file! As for ReadyBoost, I was directed by a Gizmo's response to BleepingComputer.com, which provided me a link to an article on monitoring ReadyBoost through the Administrative Tool, Performance Monitor.

This tool has showed me conclusively that the laptop is definitely benefiting from ReadyBoost, while the desktop may not be. But more testing is needed. Thank you all for your input.

 

[Digerati]

You are, right. My apologies... to all.

This all ties back to your ReadyBoost question and the 8Gb of RAM - at least on your PC. Note that ReadyBoost was introduced with Vista and one of it primary suggested uses was to move the Page File to it - the idea being it was much faster than the slow hard drives common at the time. So that's how I was diverted by the PF issue.

I think it would be a waste of money to go buy new flash drives to use as ReadyBoost devices if your system already has 8Gb. But for your notebook, since it has only 2Gb, using a flash drive for this purpose might yield noticeable performance gains - depending on how you use your notebook. But I still would recommend increasing the RAM in the notebook - if possible, for a better, faster, and permanent solution.

But since it seems you have had those flash drives "for years" - using them instead of tossing them in a drawer is probably not considered a waste of money. But I doubt it is improving your PC's performance in any significant manner. But I could be wrong.

There are all sorts of benchmarking programs - but they are all artificial because they don't reflect how you use your computer. So I would suggest for your PC, you try it without and see what happens. Just remember, it is human nature (to a fault!) to see things favorably when they are not - wishful thinking or hope are influences. So just be aware the placebo affect is very real. Conducing the test "blind" (so you don't know if the device is in use, or not) would be best.

You just don't hear about the virtues of ReadyBoost - even though it has been around for years. I think that is because the virtues didn't pan out. The USB interface just can't compete with more RAM. I suspect the faster SATA interface, and drives with much larger buffers helped too.

 

[BearPup]

Thank you for your response (and the apology). I use a RAM disk for my desktops PF. My notebook is another story.

At 2 GB, my notebook is halfway to max RAM, and to increase it I'd have to throw out the 2, 1 GB chips and then add 2, 2 GB chips. To make matters worse, the first RAM slot is not accessible from an external port cover - you have to take the case apart, then remove the keyboard, then..... In other words, it ain't going to happen!

I'm seriously thinking about taking one of the current flash drives from the desktop ReadyBoost and putting it to the notebook for that purpose. Whether I'd put my PF to that drive is more problematical, as I don't need a 16 GB PF on a 2 GB laptop! But as far as I know, I can't specify how big a PF I can dedicate on the flash drive, its an either / or proposition: either its used for a PF or its not.

At any rate, the bleepingcomputer.com site gave me the answer I was looking for (http://www.bleepingcomputer.com/tutorials/using-windows-readyboost-to-increase-performance/). Thanks to all for responding.

 

[clifford_cooley]

Okay. That's your choice but I prefer to keep an open mind.

But yet you are not keeping an open mind, when you insist on telling me I need a Pagefile when I know I do not. I'm not suggesting paging does not have its place. I am suggesting that ReadyBoost is a failed project, which is a form of paging. I am suggesting the use a Pagefile only instead of ReadyBoost assistance.

And no I did not drift from the topic. I am suggesting the use of Pagefile only when needed and the non-use of ReadyBoost because I believe it has no real benefits. It is up to the OP to make up their mind who they want to believe. If you have a difference in opinion please present them. But once again you are here attacking someone else's comment and spreading your links as if they are proof of fact.

I'm getting tired of your attacks, present your opinion if you will but stop the attacks.

 

[Digerati]

At any rate, the bleepingcomputer.com site gave me the answer I was looking for (http://www.bleepingcomputer.com/tuto...e-performance/). Thanks to all for responding.

I am glad you found your answers but please take note of the conclusion where Larry states, "you will not get the same performance increase that you will receive if you actually installed more physical RAM".

I'm getting tired of your attacks, present your opinion if you will but stop the attacks.

Where did I attack you? Someone presents a different opinion from yours and suddenly it's an attack???

But yet you are not keeping an open mind, when you insist on telling me I need a Pagefile when I know I do not.

I am willing to read and consider any evidence you present. You, however, made a point of saying you are not even going to look at the supporting evidence I provided! I provided supporting documentation because I don't expect anybody to automatically accept what I say as the Gospel, simply because I say so. "Proof of fact"??? It is called supporting evidence. You provided what? Your say so? And you know you and others don't need the PF based on what? Because your computer does not break when you disable it? What makes your system so unique that you are the exception?

I'm not suggesting paging does not have its place. I am suggesting that ReadyBoost is a failed project

I agree ReadyBoost is a failed project, but not because it is a form of paging or caching - but because it works via the slow USB port and compared to system RAM speeds, USB will create a bottleneck, not alleviate one.

I am suggesting the use a Pagefile only instead of ReadyBoost assistance

I strongly disagree with this suggestion to everyone reading here to use a PF "only instead of ReadyBoost assistance" and would ask that you PLEASE provide us some supporting evidence to a White Paper or TechNet article to support this advice you are giving.

 

[BearPup]

Yes, I'm aware that ReadyBoost's 'configured RAM' is not the same as / as good as actual, physical RAM. But from an economic standpoint, as well as a practical one for my laptop, ReadyBoost is the best answer.

 

[Digerati]

But from an economic standpoint, as well as a practical one for my laptop, ReadyBoost is the best answer.

Since you already own these devices and since many notebooks do tend to be a bit more difficult and expensive to upgrade, that makes sense.

But again, I would simply try it each way and see if you notice any difference. If you don't then I would not use ReadyBoost.

 

[clifford_cooley]

Where did I attack you? Someone presents a different opinion from yours and suddenly it's an attack???

Breaking my comment apart while you present your opinion or facts what ever they may be is attacking the comment you broke apart. I presented my comment as an opinion, there was no need in you breaking the comment apart just to present your own opinion.

My apologies to the OP for the thread turning into a rage between Digerati and myself. I promise the rages will stop. I'm out of this thread and I'm sorry for presenting my opinion, they seemed not welcome by either of you.

 

[Digerati]

Breaking my comment apart while you present your opinion or facts what ever they may be is attacking the comment you broke apart

Oh come on!

You presented an opinion about ReadyBoost, which I agreed with.

Then you "suggested" users use a PF ONLY in place of ReadyBoost (with absolutely no rationale or corroborating data to support that advice, and then you said, "if you have a difference in opinion please present them" but now have made it clear, presenting a different opinion than yours is attacking you! That's puerile.

Break apart? I was addressing specific parts because as noted, I agreed with much of what you said.

Sorry again, BearPup.

Good bye.

 

[clifford_cooley]

I don't care Digerati. In fact I've asked to be removed from Moderator standing, so I no longer have need in concerning myself with the accuracy of your comments.

 

[Digerati]

I don't care Digerati. In fact I've asked to be removed from Moderator standing

I am sorry to hear that and I think that is a loss for the site, and posters seeking advice here, Clifford, for clearly you have a lot to offer - and I mean that sincerely. Please do note we agree on much more than we don't.

so I no longer have need in concerning myself with the accuracy of your comments.

Concerned with the accuracy? That's why I post links to support my comments - so everyone can see I am not trying to blow smoke up anyone's butt - and because I know things change as technologies advance (and my aging memory fades) so constant refreshing to ensure accuracy is part of my SOP. I don't expect anyone to take what I say as the Gospel, simply because I say it.

So again, I am more than willing, and would be happy to read and consider any evidence you present that supports your claims.

But please! Just because I might present different position, that in no way is a personal attack. It sure was not meant to be, but I apologize if it appeared that way just the same.

 

[Ian]

I think this thread has run its course now, so I'm locking it now. My apologies to the OP .