Anyone tried Dirac Live 3.3.0 update?

Timmy C

Distinguished Member
Joined
Oct 10, 2003
Messages
7,973
Reaction score
1,822
Points
1,604
Age
51
Just updated mine and these blue bars seem to be a new feature but not sure what they do. Anyone else updated and had a play around yet?
 

Attachments

  • 2022-04-12.png
    2022-04-12.png
    344.7 KB · Views: 366
  • Screenshot 2022-04-12 210840.png
    Screenshot 2022-04-12 210840.png
    41.3 KB · Views: 236
Could they be the "new way to set the target curve"?
Did you check out the help via the "?"
 
What does the first line mean? Does anyone know what that actually does, as it sounds interesting?
 
What does the first line mean? Does anyone know what that actually does, as it sounds interesting?
That was what I was hoping to find out more about. I searched last night and only found the update mentioned on an Arcam and Onkyo chat thread with no detail.

When you re-load a saved project something pops up say something along the lines of optimising calculations which suggest it's doing something new. This could well be my imagination but I'm sure things sound different. I've only had time for a very quick play around though so wouldn't swear to anything.

Loading projects has also caused it to crash twice now which is new but I'm not sure that's supposed to be a feature!
 
Last edited:
I downloaded this last night and also found it very buggy.

Crashed several times then wouldn’t load the file to my NAD and I had to power cycle it or it’d get to 75% and fail.
 
Loading projects has also caused it to crash twice now which is new but I'm not sure that's supposed to be a feature!

It's definitely a new feature... customer research showed that product users like a challenge, the crashes bring more excitement and anticipation for the new software. So they prefer what the software does even more due to all the hard work installing it. 🤣
 
If you turn on and off the "detected range" checkbox do they appear and disappear? Or are they connected to the target curve at all?
 
If you turn on and off the "detected range" checkbox do they appear and disappear? Or are they connected to the target curve at all?

I've not had a chance to play with it after an initial go shortly after posting this thread. The blue bars are a new way to adjust the target curve but I didn't think to look and see if there was a help button with any guidance. Hopefully I'll be able to look again tonight.
 
Last edited:
Just had another brief play about and now gone back to the old version again. Nothing I can see in the help mentions anything about the changes.

So when you load your previous project, you see the screen as in the screenshot I first posted. I had target curve switched off at the point I took the screenshot but that would be there too otherwise. However, it's not the target curve that was used for your previously saved project. In order to view that you click on an icon up where the speaker grouping options are. You can adjust either curve using the new or old method but what is not clear to me is which of these curves it's then using to calculate.

With my original custom target curve on the screen along with the 'corrected' curve, looking just as it did previously, I hit the filter export button and Dirac did it's thing, only it doesn't sound as it did previously. I don't use Dirac in pursuit of a flat response and only use it to make minor adjustments to how things sound without it. So switching between EQ and no EQ when listening shouldn't make great differences in my case. Only now it does. Most notably a lack in the low end.

So either Dirac is working it's magic in a very different way to how it was or it's actually uploading the other target curve and not the one I'm seeing onscreen when I proceed to export. I think I'll wait to see other peoples findings before considering revisiting.
 
A little update in case anyone is interested. I re-installed the new version and it appears to be behaving itself this time and results are more in line with what I would have expected. I'm not sure why though!

A chap from Dirac posted on AVS saying that there may be problems working with old projects and suggested taking new measurements if you want to try the updated version. So far there's been no confirmation that this issue is going to be solved with a new update so either stick with what you have or get your mic out and start from scratch with the new version.
 
I had a go with 3.3.0 last night, it took a while but I was able to load my old measurements in.

I loaded the filter to my NAD T778 and did and A-B comparison between the same filter loaded from the old version.

It seems pretty much the same, except it's about 5dB quieter, so it's hard to directly compare.

I wonder if they've added some extra headroom to prevent clipping.

The auto target curve is an interesting feature, but it didn't sound anywhere near as good as my custom curve.

I'll have a go using fresh measurements at some point, it may be that it's mostly but not fully compatible with older measurements.

The project files are about a quarter of the size that they used to be though, which is nice.
 
I'll have a go using fresh measurements at some point, it may be that it's mostly but not fully compatible with older measurements.

There was an update from the Dirac guy on AVS saying you can check the conversion by comparing the measured impulse response. If it's the same as it was with the previous version then all should be good with the conversion. Or at least I assume he meant impulse response but he called it something else.

I think I'm getting on fine with it now but my amp only has the one slot so comparison isn't a quick or easy process.
 
There was an update from the Dirac guy on AVS saying you can check the conversion by comparing the measured impulse response. If it's the same as it was with the previous version then all should be good with the conversion. Or at least I assume he meant impulse response but he called it something else.

I think I'm getting on fine with it now but my amp only has the one slot so comparison isn't a quick or easy process.
I noticed last night that the old measurements loaded onto my receiver using the 3.3.0 app are wrong. Looking in the speaker configuration screen, the two filters I loaded from 3.3.0 don't have levels or distances, they're all 0dB/ms.

So this could explain why the original one was louder than the same curve loaded with the new app.

So anyone else loading old measurements with the new version should check this.

Looks like new measurements are needed for me if I want to make adjustments. When I do, I'll give the auto-target curve another try, I may not have been fair to it given it hadn't loaded correctly.
 
That’s what mine does all the time :-(
 
That’s what mine does all the time :-(
So it's not specific to 3.3.0, just a faulty load? As far as I'm aware it's the first time I've had it, but I've not gone looking for it before.

It might explain why even minor adjustments can completely ruin a "good" curve for me though.
 
So it's not specific to 3.3.0, just a faulty load? As far as I'm aware it's the first time I've had it, but I've not gone looking for it before.

It might explain why even minor adjustments can completely ruin a "good" curve for me though.

I've seen similar reported on AVS with the new version. I only have a 2 channel set up and it doesn't really effect me so I didn't pay close attention to what was being said but this note is already in the bug list at the start of their Dirac thread...

"3.3.0 Delay/gain sporadically not saved in AVR"
 
So it's not specific to 3.3.0, just a faulty load? As far as I'm aware it's the first time I've had it, but I've not gone looking for it before.

It might explain why even minor adjustments can completely ruin a "good" curve for me though.
Dirac always leaves the levels and distances set to 0. You can change them, but anything you do will be added on top of what Dirac has already done.
 
Dirac always leaves the levels and distances set to 0. You can change them, but anything you do will be added on top of what Dirac has already done.

Not with my Arcam amp it doesn't.
 
Dirac always leaves the levels and distances set to 0. You can change them, but anything you do will be added on top of what Dirac has already done.
On my NAD it always sets them to values and they're locked/greyed out.

When entering distances manually they are in meters. When Dirac has measured them they're in milliseconds.

The only adjustments you can make are during the measurements and the target curve.
 
On my NAD it always sets them to values and they're locked/greyed out.

When entering distances manually they are in meters. When Dirac has measured them they're in milliseconds.

The only adjustments you can make are during the measurements and the target curve.
On the Emotiva XMC-2, Dirac sets levels and distances to 0 after calibration. You can change them if you like, but whatever you do with be an overlay on whatever Dirac has already done. I am not sure if Dirac 3.3.0 has changed anything relative to crossover selection, but prior to 3.3.0 the crossover was set in the processor, independent of Dirac. I noticed in Jerry Austin’s paper that Dirac seems to, at least, suggest a crossover point.
 
I am not sure if Dirac 3.3.0 has changed anything relative to crossover selection, but prior to 3.3.0 the crossover was set in the processor, independent of Dirac. I noticed in Jerry Austin’s paper that Dirac seems to, at least, suggest a crossover point.

I think that's for those with the optional bass control (DLBC) add on. Something I don't think is even an option on my older Arcam model.
 
I think that's for those with the optional bass control (DLBC) add on. Something I don't think is even an option on my older Arcam model.

You're right. DLBC is not available on the older Arcam FMJ series. Also not available on NAD.

The FMJ series will show what levels and delays are applied by Dirac unless something major has been changed in v3.3.0. I've not tried this update yet personally as I really don't use Dirac nowadays.
 
I think that's for those with the optional bass control (DLBC) add on. Something I don't think is even an option on my older Arcam model.
Thanks. That makes sense. I thought I had missed it. Maybe Emotiva will have DLBC one day…
 
I tried the latest versions 3.3.2 I think and it kept crashing during the 3rd measurement on my windows PC. So I opened a ticket with Dirac about it.

In the meantime I used my previous measurement file from the earlier version and used the new version for filter design. It worked just fine. Dirac confirmed that it's ok to do this as the new version does not do measurements differently.

The new version makes it easier to set house curves too.
 
I think 3.3.3 came out yesterday. Not sure if that will solve your issue or not but worth a try,
 

The latest video from AVForums

Is 4K Blu-ray Worth It?
Subscribe to our YouTube channel
Back
Top Bottom