Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why GIMP is Inadequate

troy-sobotka.blogspot.com

Post reply on HN
Print view View original

81–90 of 108 posts

Re: Why GIMP is Inadequate

#81

Post by jerf »

DrStalker wrote: post #42
julian37 wrote: post #25

Earlier quoted context omitted.

> so if you're doing work targeted at a computer screen, it probably doesn't matter anyway Actually, it matters regardless of output device. With higher bit depths, less rounding errors accumulate over time as the image is processed, for example when effects are applied, yielding higher overall output quality. Here is an example, although after many iterations quality loss can be much more dramatic than visible in th…

> With higher bit depths, less rounding errors accumulate > over time as the image is processed, This could also be solved by another photoshop feature, adjustment layers. Basically instead of applying a filter like levels to an image you make a levels layer that applies the levels affect to everything underneath it. The original data is untouched, and you can tweak the levels at any time. I'm strictly a hobby user f…

No, that doesn't actually do anything to solve the underlying problem. Stacking five adjustment layers has the same accuracy problems as doing the same five operations sequentially. It may let you tweak them without roundtripping through further loss but the loss is already significant by that point.

Re: Why GIMP is Inadequate

#82

Post by NIL8 »

samlevine wrote: post #80
NIL8 wrote: post #79

What shocks me is the lack of competition in this field in the year 2011. Why are we still talking about choosing between GIMP or Photoshop? I know there are other programs out there that some people will claim they prefer over GIMP or PS, but for most of the planet it's GIMP or PS. Is the lack of competition due to the magnitude of such a programming endeavor or is it something else like patents? Any idea?

I've been using Pixelmator for a few months now (GIMP works great in Linux but I've found it's performance in OS X to be rather poor). It's still closed source software like PS (even if it does use open source libraries), but it's fast enough and gets the job done (my needs are really meager). The reason why GIMP doesn't have more developers is easy: Programmers don't have any itch to scratch at this point and there…

Interesting point about piracy. Without a need, there's no reason to innovate.

Re: Why GIMP is Inadequate

#83

Post by waterside81 »

It's a shame that GIMP isn't supported by more devs/money. Writing script-fus for GIMP makes it so much more useful for a developer than Photoshop. I've saved myself so much time using python-fu with GIMP to automate opening, manipulating, saving of files.

Re: Why GIMP is Inadequate

#84

Post by andrewvc »

jerf wrote: post #81
DrStalker wrote: post #42

Earlier quoted context omitted.

> With higher bit depths, less rounding errors accumulate > over time as the image is processed, This could also be solved by another photoshop feature, adjustment layers. Basically instead of applying a filter like levels to an image you make a levels layer that applies the levels affect to everything underneath it. The original data is untouched, and you can tweak the levels at any time. I'm strictly a hobby user f…

No, that doesn't actually do anything to solve the underlying problem. Stacking five adjustment layers has the same accuracy problems as doing the same five operations sequentially. It may let you tweak them without roundtripping through further loss but the loss is already significant by that point.

Technically yes, but not at all in practice. Once you run 3 adjustments you generally want to tweak the first. Without adjustment layers you usually just run a fourth transformation.

Re: Why GIMP is Inadequate

#85

Post by andrewvc »

jerf wrote: post #81
DrStalker wrote: post #42

Earlier quoted context omitted.

> With higher bit depths, less rounding errors accumulate > over time as the image is processed, This could also be solved by another photoshop feature, adjustment layers. Basically instead of applying a filter like levels to an image you make a levels layer that applies the levels affect to everything underneath it. The original data is untouched, and you can tweak the levels at any time. I'm strictly a hobby user f…

No, that doesn't actually do anything to solve the underlying problem. Stacking five adjustment layers has the same accuracy problems as doing the same five operations sequentially. It may let you tweak them without roundtripping through further loss but the loss is already significant by that point.

Plus selectively painting the mask of an adjustment layer is useful as well

Re: Why GIMP is Inadequate

#86

Post by aw3c2 »

yesimahuman wrote: post #75
joakin wrote: post #9

I expected this to be a rant or flame, but in fact he has described wonderfully what needs to be said. Besides from all this pro features that GIMP lacks, in my opinion its biggest flaw its the UI. Its poor and raw, and makes sense if you are a programmer... Mainly this is what keeps it far from the regular users I know. Hope It doesnt die, its a very good open source multiplatform editing tool...

I honestly don't see what the big deal is. Comparing GIMP usability to Photoshop usability only makes sense if you have no experience with either . Many people say GIMP is difficult yet forget they had a hard time learning Photoshop to begin with. I personally find Photoshop daunting and GIMP pretty easy to use for most of what I need.

My dad is 70, likes to rant about broken/bad software and recently discovered GIMP for photo manipulation. He is very happy and has not said a single bad thing (which actually surprised me).

Re: Why GIMP is Inadequate

#87

Post by jerf »

andrewvc wrote: post #84
jerf wrote: post #81

Earlier quoted context omitted.

No, that doesn't actually do anything to solve the underlying problem. Stacking five adjustment layers has the same accuracy problems as doing the same five operations sequentially. It may let you tweak them without roundtripping through further loss but the loss is already significant by that point.

Technically yes, but not at all in practice. Once you run 3 adjustments you generally want to tweak the first. Without adjustment layers you usually just run a fourth transformation.

Ye gods, can't you read until my third sentence between smashing the reply button and "correcting" me?

Re: Why GIMP is Inadequate

#88

Post by bitwize »

The problem is that the intersection of fosstards and serious print designers is vanishingly small. If you're doing pro press work, you're going to buy the tool that everybody else uses and that print shops have standardized upon. That tool is Photoshop.

Re: Why GIMP is Inadequate

#89

Post by krakensden »

billhasmail wrote: post #15

"Important progress towards high bit-depth and non-destructive editing in GIMP has been made. Most color operations in GIMP are now ported to the powerful graph based image processing framework GEGL, meaning that the internal processing is being done in 32bit floating point linear light RGBA. By default the legacy 8bit code paths are still used, but a curious user can turn on the use of GEGL for the color operations…

GEGL is not fully baked yet, using it reduces stability and the operations you can perform. Reviewing the default, non-GEGL GIMP was the right decision.

Re: Why GIMP is Inadequate

#90

Post by aphyr »

jerf wrote: post #81
DrStalker wrote: post #42

Earlier quoted context omitted.

> With higher bit depths, less rounding errors accumulate > over time as the image is processed, This could also be solved by another photoshop feature, adjustment layers. Basically instead of applying a filter like levels to an image you make a levels layer that applies the levels affect to everything underneath it. The original data is untouched, and you can tweak the levels at any time. I'm strictly a hobby user f…

No, that doesn't actually do anything to solve the underlying problem. Stacking five adjustment layers has the same accuracy problems as doing the same five operations sequentially. It may let you tweak them without roundtripping through further loss but the loss is already significant by that point.

Er, theoretically, no. I don't know what Photoshop's implementation is, but computing the cumulative effect of the adjustment layers will result in less error than applying them in sequence. That's part of why GEGL is being written.

[Edit] I'm referring to the fact that fixed-precision arithmetic loses (well, stochastically speaking) precision with each operation. You can condense the layer effects and apply the 8-bit operation only once to minimize error. This is especially true if your layer combinator works at a higher level of precision than the layer operation itself.

[Edit 2] As an example, consider the layer operations "Multiply each channel by 0.5" and "Multiply each channel by 2." Applying these layers separately to a value loses something like a half bit of precision (more for smaller values, less for larger ones). A layer combinator could multiple 0.5 by 2 and yield ~1.0 (in general, subject to the floating point precision which is generally finer than the image precision). That results in a smaller error.

Post reply on HN