You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,
Nice wrapper, thanks!
Looks like I need some help with understanding "[demand-driven feature]( https://github.com/libvips/libvips/wiki/Why-is-libvips-quick)".
As far as I understand libvips can process image without loading it to RAM fully.
For Jvips it is possible to pass MappedByteBuffer as input; But how about output?
Does it mean JVips doesn't support this feature or I missed the point of "demand-driven" and it applies only for reading.
Thanks.
B.
The text was updated successfully, but these errors were encountered:
Yes, we didn't implement it because when we measured cpu performance between mapped and byte array, the results were very similar.
Thus, we decided to keep a straightforward implementation for our app.
We didn't measure the impact on the memory, it would be a nice to have.
If we implement this feature, we must provide benchmark on the memory consumption.
Hi,
Nice wrapper, thanks!
Looks like I need some help with understanding "
[demand-driven feature]( https://github.com/libvips/libvips/wiki/Why-is-libvips-quick)
".As far as I understand libvips can process image without loading it to RAM fully.
For Jvips it is possible to pass MappedByteBuffer as input; But how about output?
Does it mean JVips doesn't support this feature or I missed the point of "demand-driven" and it applies only for reading.
Thanks.
B.
The text was updated successfully, but these errors were encountered: