Tony-Trim Anyone??

O

Old Hippie

Supreme [H]ardness
Joined
Oct 31, 2005
Messages
6,013
Has anyone played with this Tony-Trim procedure?

I used it on my Intel G1 RAID0 array and went from 472 MB/sec to 502 MB/sec after the treatment.

Anybody else played with this FreeSpace cleaner or Tony-Trim thing?
 
No one has tried this?

I guess the "Hard" SSD faction of this forum could use a little viagra? :D
 
I thought about running the procedure on my 60GB OCZ Vertex, however my read/write speeds are practically perfect right out of the box. (running the drive for about 4 days now, compared to the marketed read/write speeds)
 
Acejam said:
I thought about running the procedure on my 60GB OCZ Vertex, however my read/write speeds are practically perfect right out of the box. (running the drive for about 4 days now, compared to the marketed read/write speeds)
Click to expand...

The "out of box" speeds will slow as you hit steady state when all the blocks have been written at least once.

If you kept your out of the box drive testing graphs, it's interesting to see the results after a month of usage.

There's been soo much moaning and groaning about the Intel G1 drives not getting TRIM and TRIM not working with RAID arrays that I thought there'd be a little more interest in this solution.

I'm kinda thinkin' it's the same old stuff....people would just rather bitch and cry than actually try anything that's not handed to them on a silver platter.
 
I will check into this as my ssd has no firmware or trim and i will be getting a second drive for raid, thank you for a possible solution for me as i bought the drive on impulse not research.
 
do you think you will actually see a slowdown of the drives in as little as a month?
 
So has anyone had a chance to put "TONY-TRIM" through its paces? From what I've read there have been mixed results from using it. I'm wondering what you all think.
 
Naturally I have, and it works great for me.

I've only read one negative report and would appreciate other links where it didn't work.
 
Old Hippie said:
Naturally I have, and it works great for me.

I've only read one negative report and would appreciate other links where it didn't work.
Click to expand...

I came across a few while combing through Google. Basically people reporting that performance was lower after running it. That is why I posted something here - I'd like to hear from the [H] community directly. Thanks.
 
i have a G1 160 that i will do this with. i am only 2 weeks in to a fresh format so ill probably try it around xmas time
 
Just for some explanation, PerfectDisk is compacting the data into the fewest number of blocks, to give you the most completely free blocks possible. AS Cleaner is then filling up all the free space with a file containing "1" bits, which correspond to "empty". The file is then deleted, marking all those bits as available again. Even though those bits won't actually get cleaned by TRIM in unsupported setups, the data that was previously in the file is the equivalent of "empty".

Due to the controller's automatic remapping of file locations and stuff, I'm not sure how well PerfectDisk would actually do at compacting the data. I suppose it could try to bunch files together and force them to be rewritten, hopefully all getting written to a single block. However, with all the smarts in the controller itself, I can see this easily getting completely undone. *shrug*

The AS Cleaner step makes sense, but I don't know enough about the underlying technology to know for sure if it would actually work on all SSDs. Just like there's a difference between NULL and 0, there may be a difference between writing a 1 to a bit, and that bit being erased by the controller (even if it does result in the bit showing 1). It should be easy enough to test this though - just use the drive enough to cause degradation, then run AS Cleaner and see if it helps. If it's really that simple, then any free space wiper that's capable of writing 1's (FF) should work fine for this.
 
You must log in or register to reply here.
Back
Top