tomgalileo

I am experience a very low accuracy of timepoints.

I want a timepoint to create a loop in the transport, resetting 

transport's ticks to 0., every time timepoint sends out its bang.

Well this bang sometimes does not come out, especially if you 

Hello
I am experience a very low accuracy of timepoints.

I want a timepoint to create a loop in the transport, resetting  
transport's ticks to 0., every time timepoint sends out its bang.
Well this bang sometimes does not come out, especially if you  
increase the tempo speed.
Is there a way to work around this?

max5-timepoint-accuracy

Show us the patch.We'll have a look at it.

hello, here's my patch. I supposed I was obliged to post it through forum, so it took me a while as I forgot password and so on...

Why are you using different named transports?

If they're not essential to your patch, then maybe it will work better just off the same transport.

Your patch seems to work ok if I just leave it alone. Are you saying that the problem lies in timepoints skipping while "manually" changing tempo?

Well the different names were due because there were a lot more 

transports in the real patch, but I have deleted to show you a 

2) after changing a tempo (to a faster one), it has a good behaviour 

3) i am using max5 together with max 4.6.3 sending bangs from 

timepoints via udpsend/receive: when you simpli click on a patch on 

max 4.6 to switch from max5 view, if there were a timepoint at that 

time, it skips. Maybe this is not duedirectly to timepoints.

On 17 Jul 2008, at 23:41, Andrew Pask wrote:

> Why are you using different named transports?

> If they're not essential to your patch, then maybe it will work 

> Your patch seems to work ok if I just leave it alone. Are you 

> saying that the problem lies in timepoints skipping while 

Well the different names were due because there were a lot more  
transports in the real patch, but I have deleted to show you a  
simplified one.

The skipping happens:
1) when changing the tempo.
2) after changing a tempo (to a faster one),  it has a good behaviour  
a couple of times, then skips.
3) i am using max5 together with max 4.6.3 sending bangs from  
timepoints via udpsend/receive: when you simpli click on a patch on  
max 4.6 to switch from max5 view, if there were a timepoint at that  
time, it skips. Maybe this is not duedirectly to timepoints.

>
> Why are you using different named transports?
>
> If they're not essential to your patch, then maybe it will work  
> better just off the same transport.
>
> Your patch seems to work ok if I just leave it alone. Are you  
> saying that the problem lies in timepoints skipping while  
> "manually" changing tempo?
>
>
> -A
>
>


I was able to recreate the problem you are seeing. I'm going to log it but please don't expect a quick fix. I would say it's an area of ITM performance we could improve on. In the meantime, changing tempo does not seem to goof with timepoints in the same transport.

Actually I think I take that back. It seems like named transports other than the main one are missing timepoints like crazy.

Thanx Andrew! No hurry for me, I use max 5 just to learn and try 

On 18 Jul 2008, at 17:50, Andrew Pask wrote:

> I was able to recreate the problem you are seeing. I'm going to log 

> it but please don't expect a quick fix. I would say it's an area of 

> ITM performance we could improve on. In the meantime, changing 

> tempo does not seem to goof with timepoints in the same transport.

Thanx Andrew! No hurry for me, I use max 5 just to learn and try  
things on it by now.
ciao!

>
>
>
> I was able to recreate the problem you are seeing. I'm going to log  
> it but please don't expect a quick fix. I would say it's an area of  
> ITM performance we could improve on. In the meantime, changing  
> tempo does not seem to goof with timepoints in the same transport.
>
>
>
>
> Cheers
>
>
> -A
>


maybe the resetting function can be in the future embedded so that 

there will be no more need to do the bang 0. to itself...

which is good but a bit not so exceptional i think...

On 18 Jul 2008, at 19:43, Andrew Pask wrote:

yes please! as is is a very cool thing!
maybe the resetting function can be in the future embedded so that  
there will be no more need to do the bang 0. to itself...
which is good but a bit not so exceptional i think...

>
>
>
> Ok - this should be improved in 5.04
>
> -A


max5: timepoint accuracy