Export thread

  • Be sure to checkout “Tips & Tricks”
    Dear Guest Visitor → Once you register and log-in:

    This forum does not automatically send notices of new content. So if, for example, you would like to be notified by mail when Steve posts an update to his blog (or of any other specific activity anywhere else), you need to tell the system what to “Watch” for you. Please checkout the “Tips & Tricks” page for details about that... and other tips!

    /Steve.

Progress bar resetting

#1

D

Dagannoth

I have a strange little mSATA drive (32 GB Emphase G5RM3G032-M) I bought from eBay that seems to have some problems. The problems are severe enough that it occasionally freaks out ReadSpeed. I finally got a capture card, and the first run I captured with it couldn't have been better. The previous runs where I noticed this problem were much more subtle. The resetting begins at the 50% test.



#2

P

PHolder

Well @Steve would have to reply, as it's his code, but to me it looks like something is happening to interrupt the reading (an error maybe) and it restarts, or something.


#3

D

DanR

Hmm . . . Googling the drive model gives me this:

G5RM3G032-M
Emphase mSATA MLC solid state drive with read/write speeds of 470/365 MB/sec.

ReadSpeed results:
0%: stutters a bit but does one pass
25%: fast, smooth, one pass
50%: takes 5 tries
75%: takes 2 tries
100%: fast smooth, one pass

Definitely looks like read errors at 0%, 50%. and 75%. The apparent speeds at 50% and 100% seem too fast; 0% and 75% too slow; and 25% just about right?

I'm thinking this is a candidate for SpinRite Level 3?


#4

D

Dagannoth

Definitely looks like read errors at 0%, 50%. and 75%. The apparent speeds at 50% and 100% seem too fast; 0% and 75% too slow; and 25% just about right?

The resetting usually takes place at 0%, and also ends up being around 155 MB/s. This drive is very inconsistent. I did the sub-block tests on it, but didn't include it in that post because the results fluctuated too much.

Code:
155.2    436.0    542.9    448.3    542.8
154.4    435.6    542.8    448.1    542.7
397.8    439.3    542.9    448.1    542.8
475.0    436.3    542.9    449.5    542.8
154.3    435.2    542.8    448.0    542.7
154.2    435.3    542.8    448.1    542.7
155.4    441.4    540.6    448.0    542.2
155.4    441.8    542.5    447.8    542.3
155.5    442.7    542.5    448.0    542.3
156.8    442.9    541.9    447.9    541.6
156.6    443.3    541.8    447.6    541.4
156.2    442.7    541.9    445.2    541.5
389.2    441.3    540.7    446.1    540.2
156.7    441.2    540.6    446.7    539.6
530.3    441.4    540.5    446.4    539.8

I included the logs for those results below, as well as a couple more runs from a few days earlier, and the smartctl output.

I'm thinking this is a candidate for SpinRite Level 3?

I couldn't agree more, but I want to save it for 6.1 :)

Attachments


  • Emphase G5RM3G032-M.zip
    38.1 KB · Views: 57

#5

D

DanR

There is definitely something not right with that drive. If the BIOS has an ATA controller option, then SpinRite 6.0 would work. Otherwise waiting for 6.1 is the best bet.


#6

D

Dagannoth

It's a SATA drive, so 6.0 would definitely work on it. It's a drive I bought just for testing, so it doesn't have any data on it. I figure a known bad drive is a better test for 6.1 than a known good drive.


#7

D

DanR

Agreed! Plus, 6.1 would be quite a bit faster than 6.0. It's an excellent candidate for 6.1.


#8

Steve

Steve

Gang: Finally dropped by here to catch up. (Checking in on “ReadSpeed Problems” is my "go to".)

What's happening is that ReadSpeed is encountering a hard error (unreadable sector) from the drive. So it aborts the transfer, moves 1GB inward (or outward if testing the 100% point), resets the bar graph and tries again. In other words, on a sufficiently troubled drive, this behavior is expected and I designed for it. (y)


#9

D

Dagannoth

Ah, good to know it's not a bug :). Does ReadSpeed log this behavior anywhere, or would that be a feature request?


#10

P

PHolder

this behavior is expected and I designed for it
Is this documented anywhere else but in the message I quoted? Perhaps it should be mentioned on the GRC page.


#11

D

Dagannoth

Also, is it possible to turn off that retry and have it return 'FAILURE' instead? I think the vastly different numbers this drive is returning could be from reading different areas.