I've not been able to repro with that script on windows today Igor. Sorry I
can't be of more help.
On 15 September 2016 at 19:51, Frank Harrison <fr...@thefoundry.co.uk>
> No probs Igor,
> unfortunately I've been unable to repro it on my mac. I am getting mem
> alloc issues and it's falling back to the CPU path on my GeForce GT 750M,
> so that's probably why.
> My windows build for Kronos is still going (don't ask!) so I've not been
> able to try and repro there. Given that it's taken so long, I'll try and
> have a look in the morning and I'll point the guys who know Kronos
> backwards this way.
> Looking at your script you mention the GTX980. I know that our support
> guys have fielded a few questions with that GPU so you might actually be
> better off contacting them - ask for Pete Coleman as he and I were looking
> into this briefly some time ago IIRC the GPU name.
> As for the bounds issue, perhaps try and ensure that all inputs have the
> same-same format, and/or make sure you clear you cache.
> Sorry I've not been of more help (yet?).
> On 15 September 2016 at 19:04, Igor Majdandzic <subscripti...@badgerfx.com
> > wrote:
>> Even when I type this in the cmd, I dont get an error report
>> C:\Program Files\Nuke10.0v4>Nuke10.0.exe --crashhandling 1 -F 1021-1022
>> -x myFile.nk
>> But someone suggest to check the frame values, and indeed I managed to
>> render the two frames . I truncated after two decimals and I was able to
>> If that was the error I dont know. But it worked.
>> None the less, Frank thank you. If oyu find something I am all ears why
>> that weird Different Bounds Error was occuring.
>> Cheers, the drink's on me.
>> Am 15.09.2016 um 19:36 schrieb Frank Harrison:
>> On 15 September 2016 at 18:20, Daniel Hartlehnert <dah...@gmx.de> wrote:
>>> About that crash reporting:
>>> I used to get alot of crashes without crash reporter coming up. It
>>> wasn’t Nuke10 but some versions ago, maybe 8. Also, almost no shop i ever
>>> worked at allows for outside communication of the workstation nuke is
>>> running on. IT security reasons whatsoever. So even if there is a report,
>>> it cannot be sent.
>> Yep, but we have a set of tools that, if the studio are willing to work
>> with us, will allow them to send us such crash reports - although these are
>> still in alpha.
>>> I thought this is worth mentioning. Don’t be fooled by the numbers of
>>> reports coming in. There are much more crashes than reports.
>> This is our tacit assumption :) No reporting tool will ever be perfect,
>> but we can try to make it better.
>> We introduced the crash-reports in Nuke 8.0 (IIRC) and we have seen quite
>> an improvement in the reports since then. To back that up we regularly seek
>> out other feedback and that seems to back it up - that it's gotten much
>> better. Feel free to disagree (and *please* do).
>> [image: CHIMNEY]
>> *Igor Majdandzic*
>> Compositing Supervisor
>> CHIMNEY Deutschland GmbH & Co. KG
>> Hanauer Landstraße 196
>> 60314 Frankfurt am Main
>> Direct: +49 69 989 56 431 - *NEW PHONE NUMBER* -
>> Fax: +49 69 904 36 68 29
>> Mobile: +49 176 321 375 72
>> Berlin, Frankfurt/Main, Stockholm, Malmö, Warsaw, Copenhagen, Sydney,
>> AG Charlottenburg HRA 44507 B; Pers. haft. Gesellschafter: Chimney
>> Deutschland GmbH
>> AG Charlottenburg HRB 130013 B; Geschäftsführer: Ralf Drechsler, Michal
>> Kalinowski, Carl Henric Larsson
>> Nuke-users mailing list
>> Nukeemail@example.com, http://forums.thefoundry.co.uk/
> Frank Harrison
> Senior Nuke Software Engineer
> The Foundry
> Tel: +44 (0)20 7968 6828 - Fax: +44 (0)20 7930 8906
> Web: www.thefoundry.co.uk
> Email: frank.harri...@thefoundry.co.uk
Senior Nuke Software Engineer
Tel: +44 (0)20 7968 6828 - Fax: +44 (0)20 7930 8906
Nuke-users mailing list