Using FUA cache clearing

[sredmyer]

When I run the FUA test I get results showing that (if I am interpreting the results correctly) it takes MUCH longer to clear with FUA than without.



This shows that the test took only 1735 ms when FUA is not used but took 29234 ms when using FUA. This seems to be saying that using FUA will increase the time required to clear the cache nearly 17 fold!

Am I interpreting this wrong or does this indicate that FUA should not be used where disc throughput is a concern. Further are the non-FUA means of cache clearing as effective as FUA?

Thanks,

[EliC]

Re: Using FUA cache clearing

My understanding is that the FUA command only works with real plextor drives, though many non-plextor drives I have tested with dBpoweramp's test indicate that the FUA command works, I don't trust the results.

I don't know if other makers use the FUA command or how reliable the test really is.

[sredmyer]

Re: Using FUA cache clearing

My understanding is that the FUA command only works with real plextor drives, though many non-plextor drives I have tested with dBpoweramp's test indicate that the FUA command works, I don't trust the results.

I don't know if other makers use the FUA command or how reliable the test really is.

This is a "real Plextor". It is a Plextor PX-708A

[Spoon]

Re: Using FUA cache clearing

For your drive FUA works.

[EliC]

Re: Using FUA cache clearing

When I run the FUA test I get results showing that (if I am interpreting the results correctly) it takes MUCH longer to clear with FUA than without.



This shows that the test took only 1735 ms when FUA is not used but took 29234 ms when using FUA. This seems to be saying that using FUA will increase the time required to clear the cache nearly 17 fold!

Am I interpreting this wrong or does this indicate that FUA should not be used where disc throughput is a concern. Further are the non-FUA means of cache clearing as effective as FUA?

Thanks,

If I interpret it correctly, the time is so much shorter without FUA because the cache is read. The time is much longer with FUA because the cache is not used and the drive has to re-read the disc.

[sredmyer]

Re: Using FUA cache clearing

If I interpret it correctly, the time is so much shorter without FUA because the cache is read. The time is much longer with FUA because the cache is not used and the drive has to re-read the disc.

So to answer my question then, is it better to use FUA or not when disc throughput is a concern. If it is better than I guess I still do not understand the interpretastion of the results.

[EliC]

Re: Using FUA cache clearing

It is better.

It take longer with FUA because without FUA the drive reads the data from the cache, not the disc. You don't want the data from the cache, it will match the first data, because it is the same data. You want the drive to re-read the disc, which takes longer, but is the whole point.

[sredmyer]

Re: Using FUA cache clearing

Spoon,

I am still not sure I understand this feature. Another member on the forum has stated that he thought the feature did not clear the cache but rather disabled it.

The text in the CD Ripper configuration and in the message that comes up, when the "Test FUA Support" button is clicked, clearly state that the FUA feature actually clears the cache (note the highlighted text).



Further the help file has this to say on the issue.

Clear Read Cache with FUA (dBpoweramp Reference required) certain Plextor drives can quickly remove the cached audio using a command called FUA, if supported a previously detected Drive Read Cache should show 0 KB re-running Detect with FUA checked

Based on these texts, I have been under the impression that the reason for using FUA is to quickly clear the cache. If the FUA feature were not used the software would need to read other data into the cache to defeat it rather than simply clear it. Overwriting the cache this way would obviously have a negative impact on overall ripping speed since it would require reading more actual data from the disc.

However if my interpretation is correct then the result of the FUA test makes no sense.



As you can see, the text of this message states:
"FUA appears to clear the drive cache and will be used."
"Test time without FUA: 1735ms Test time with FUA: 29234ms"

This tells me that when dbp cleared the cache by overwriting (not using FUA) it took 1735ms to do so. However when dbp cleared the cache using the FUA feature it took 29234ms.

So my questions are:
First given all the text above, how can my interpretation that FUA clears the cache be wrong?
Second if in fact my interpretation of the FUA feature is correct then, given these numbers why should this feature be ever be used?

[Spoon]

Re: Using FUA cache clearing

No, you would expect the FUA reported test time to be much larger, because if it is actively clearing the cache then a re-read (of the test) has to come from the Disc. The Without FUA test time is using the cache so is very low (the cd only reads once).

[sredmyer]

Re: Using FUA cache clearing

No, you would expect the FUA reported test time to be much larger, because if it is actively clearing the cache then a re-read (of the test) has to come from the Disc. The Without FUA test time is using the cache so is very low (the cd only reads once).

Ah I get it (light bulb goes off over head :smile2. The test is reporting how long it takes to read (multiple times) a certain bit of data not how long it took to defeat the cache.

Thanks Spoon

[larrytru]

Re: Using FUA cache clearing

Testing my HL-DL-ST GSA-T10N drive reports that FUA works on it.The time tken with FUA is much longer. But I've seen many warnings not to use it on non-Plextor drives. If its working, why not use it? Are the warnings left over form a time when this feture was not reliable on other manufacturers drives or is that still the case.

Thanks

Larry

[Spoon]

Re: Using FUA cache clearing

I personally would only enable that option for Plextors.

[timreczek]

Re: Using FUA cache clearing

Sorry to revive a dead thread, but I have a new Samsung SH-B083L BD/DVD drive that I'm trying to set up.

CD Ripper reports that the drive supports FUA, and cachex does as well.

cachex output:
Drive on D is SAMSUNG DVDWBD SH-B083L SB00

[+] Buffer size: 2048 kB, read cache is enabled
[+] Supported read commands: BEh A8h(FUA) 28h(FUA) D4h(FUA) D5h(FUA) D8h(FUA)
[+] Plextor flush command: accepted
[+] Plextor flush tests: 5/5

@Spoon - do you still recommend not using FUA for all non-Plextor drives?
If I use FUA and it is not properly supported / implemeted by the drive, what would the symptoms be?

Thanks,
Tim

[Spoon]

Re: Using FUA cache clearing

I would not use FUA on anything except a plextor.

[Nescio67]

Re: Using FUA cache clearing

I would not use FUA on anything except a plextor.

"Where can I enable the FUA option? I've a Plextor PX-760A"
Sorry the issue is already solved, forgotten that this was part of the configuration:yawn: