Text Editor Performance Tests

Discuss anything related to portable freeware here.
Message
Author
TP109
Posts: 560
Joined: Sat Apr 08, 2006 7:12 pm
Location: Midwestern US

Re: Text Editor Performance Tests

#106 Post by TP109 » Wed Jan 27, 2016 9:57 am

Tested the following editors:
Editbone v10.2.1
SynWrite v6.20.2185


Test results uploaded to below link:
http://s000.tinyupload.com/index.php?fi ... 1714834197

TP109
Posts: 560
Joined: Sat Apr 08, 2006 7:12 pm
Location: Midwestern US

Re: Text Editor Performance Tests

#107 Post by TP109 » Wed Jan 27, 2016 10:30 am

Tested the following editors:
Editbone v10.3.1

Found a small error on previous results which was corrected.

Test results uploaded to below link:
http://s000.tinyupload.com/index.php?fi ... 6151962963

TP109
Posts: 560
Joined: Sat Apr 08, 2006 7:12 pm
Location: Midwestern US

Re: Text Editor Performance Tests

#108 Post by TP109 » Tue Feb 02, 2016 2:24 pm

I'm not going to test anymore editors with XP. The next series of tests will be performed on a Win7 system. Although the XP speed test results are probably not very accurate, they can be helpful for making comparisions. The RAM, VM and max file size measurements are likely more accurate across systems. In any case, I now have a defined set of editors to test. Future testing will be refined based on what I have learned. Not sure when I will get started, but within a few weeks I think.

TP109
Posts: 560
Joined: Sat Apr 08, 2006 7:12 pm
Location: Midwestern US

Re: Text Editor Performance Tests

#109 Post by TP109 » Tue Feb 09, 2016 4:18 pm

Although no longer testing performance on XP, I'm still using XP for reliability/stability testing. In the end, I will probably create 3 categories: 1.very stable, 2.stable, and 3.mostly stable, or something similar. Here is an updated list:

These are the editors that have never crashed, locked up, froze etc. during testing:
BVD NotePad 5.2
Editor2 2.9.1.13 Ansi
EverEdit 2.9.0.2074
NFOPad 1.68n
SSL Program Edit 3.0
TopGun 2.4
Win32pad 1.5.10.4n
X-ConTEXT 0.98.6 rev3

These editors occasionally lockup or crash:
AkelPad 4.9.2
EditPadLite7 7.4
EditPadLite7 7.3.7
Notepad++ 6.8.8
Notepad2-mod 4.2.25 r964
SynPad 1.2.11.69

These editors have crashed, locked up, and so forth - more so than those listed above:
AEdit 4.0-SX r3
CudaText 1.2.12.0
Editbone 9.3.4
Editbone 10.3.1
Editbone 10.2.1
Editbone 10.1.1
Extreme Editor 7.1.2.5
Geany Portable 1.24
Geany Portable 1.22
Kudaz 2.2.1.2
MadEdit 2.9.1
MadEdit-mod 0.3.9
Minimum Profit (GUI) 5.2.10
nPad2 3.1.3.36
PlainEdit (non-Net ver) 1.7.6.0 (non-NET)
QuickEditor 3.5
SynWrite 6.19.2150
SynWrite 6.16.2010
X-Scite 3.5.1

TP109
Posts: 560
Joined: Sat Apr 08, 2006 7:12 pm
Location: Midwestern US

Re: Text Editor Performance Tests

#110 Post by TP109 » Wed Feb 10, 2016 3:46 pm

Tested Notepad2 v4.22.5 Bookmark Edition R4.

Results uploaded to:
http://s000.tinyupload.com/index.php?fi ... 5550197208

This should complete the XP performance tests. Next series of tests will be using Win7. There will be changes for the Win7 tests because it's going to include more parameters. For instance, peak RAM usage and possibly total installation size and more.

Also, test file load times will be refined although that hasn't been defined 100% yet. Load times are tricky because the file loading process and what the user perceives differs among the various text editors. For instance, some text editors have a frozen UI while loading files, while others have a usable interface. Other editors don't display anything while a file loads while others display the beginning lines of the file even though the file hasn't fully loaded. In addition, some editors provide a progress indicator during load while others don't, which can make it difficult to determine if the text editor is actually doing anything at all.

Usually, the difference between load time and initialization time is a non-issue for empty launches and small files, but this becomes more complex for large test files. The XP tests tried to determine what is most useful to the user. IMO the most important feature for the typical user is to know what is contained in a file, i.e., if they are opening the correct file or not, and secondly, whether the file will load at all and if so, how long it will take. For most editors the load times for the XP tests really don't indicate a fully loaded file, only an apparently loaded file, meaning that the user can identify if the correct file is being loaded (entire file usually isn't editable). That is why an editor like EditPadLite7 appears to be so quick for loading test files. In the case of EditPadLite7, the UI is usable and the partial contents of the file are displayed even though the entire file hasn't actually loaded. But because the user can identify the file, interact with the UI, or even cancel or change the process, it's defined as loaded for the purposes of these tests. That's probably going to change for the Win7 tests where UI initialization and file load times will be measured as separate parameters. There are some technical issues involved and this may not be possible for all the editors previously tested. I'm still in early stages of this however, so things may change as I go forward.

edited 2/10/2016 7:14PM
Some new parameters I'm likely to include in the next series of tests:
Peak RAM usage
UI Initialization time
Load completion time
Standard deviation or Initialization/Launch & Load Time variation - A number of editors often appear to launch with an inconsistent "lag" or "hang" which wasn't always taken into account for the XP tests since these tests attempted to evaluate overall responsiveness. The next phase of testing will be more demanding. This lag is why a fast editor may launch slowly at times. I need to be careful about this measurement however to ensure the lag measurements are repeatable, since any individual lag occurrence could be related to the state of the system.
Last edited by TP109 on Thu Feb 11, 2016 10:08 pm, edited 6 times in total.

User avatar
tactictoe
Posts: 283
Joined: Thu Dec 10, 2015 10:56 am
Location: A galaxy far far downunder
Contact:

Re: Text Editor Performance Tests

#111 Post by tactictoe » Wed Feb 10, 2016 10:19 pm

:wink: Very cool Test, Thanks for sharing,
“You never fail until you stop trying.”

TP109
Posts: 560
Joined: Sat Apr 08, 2006 7:12 pm
Location: Midwestern US

Re: Text Editor Performance Tests

#112 Post by TP109 » Thu Feb 11, 2016 11:47 am

Thanks,

If you have any suggestions, let me know.

User avatar
tactictoe
Posts: 283
Joined: Thu Dec 10, 2015 10:56 am
Location: A galaxy far far downunder
Contact:

Re: Text Editor Performance Tests

#113 Post by tactictoe » Thu Feb 11, 2016 8:46 pm

With a result like this:
TP109 fantastic test result on text editor - Great work.png
I do not know what to suggest... It's complete and professional.
All I can say here, except I was gobsmacked, is: What a fantastic Job you have done here TP109. That is more than helping and you probably invested lots of time to do this report.
I can not talk for the whole community but It could be good to acknowledge such effort. Thank you so much.
On the other hand if you do lots of test like this I might come handy. How? I could write a software CSV database type like that would display this sort of result and of course have all the facilities of a regular database (edition, input, querry, export to xls, etc.). Let me know.

:) Have a very special nice day TP109. :)
“You never fail until you stop trying.”

TP109
Posts: 560
Joined: Sat Apr 08, 2006 7:12 pm
Location: Midwestern US

Re: Text Editor Performance Tests

#114 Post by TP109 » Thu Feb 11, 2016 10:31 pm

Thanks, but there's always room for improvement. A database may be beneficial, depending on how much additional data is collected. I deliberately kept the number of parameters down for the XP tests to make the presentation of the results easier to comprehend. Once I have all the parameters identified for the next test series, it could be better have that in a database. Although I already had something in mind, your offer provides another option. Right now, I am in the planning stages, but I will certainly let you know which direction I will take.

User avatar
tactictoe
Posts: 283
Joined: Thu Dec 10, 2015 10:56 am
Location: A galaxy far far downunder
Contact:

Re: Text Editor Performance Tests

#115 Post by tactictoe » Fri Feb 12, 2016 2:04 am

TP109 wrote:Thanks, but there's always room for improvement. A database may be beneficial, depending on how much additional data is collected. I deliberately kept the number of parameters down for the XP tests to make the presentation of the results easier to comprehend. Once I have all the parameters identified for the next test series, it could be better have that in a database. Although I already had something in mind, your offer provides another option. Right now, I am in the planning stages, but I will certainly let you know which direction I will take.
Okay then, :D
Have a nice day.
“You never fail until you stop trying.”

User avatar
Midas
Posts: 5522
Joined: Mon Dec 07, 2009 7:09 am
Location: Sol3

Re: Text Editor Performance Tests

#116 Post by Midas » Fri Feb 12, 2016 6:34 am

FYI, here's a rather comprehensive list of Notepad alternatives I just found:
Plus a couple of suggestions garnered from Stack Overflow: PilotEdit Lite (non-portable*) and Glogg (non-portable, see http://www.portablefreeware.com/forums/ ... hp?t=15251). Some people over there advocate using hex editors to edit very large files -- like HxD or non-portable HexEdit(§). YMMV.
  • *) Saves settings to '%APPDATA%\PilotEdit' and 'HKCU\Software\PilotEdit\PilotEdit';
    §) Saves settings to 'HKCU\Software\Catch22\HexEdit 2.0 beta5'...

TP109
Posts: 560
Joined: Sat Apr 08, 2006 7:12 pm
Location: Midwestern US

Re: Text Editor Performance Tests

#117 Post by TP109 » Fri Feb 12, 2016 11:07 am

Midas wrote:FYI, here's a rather comprehensive list of Notepad alternatives I just found:

http://listoffreeware.com/list-best-free-notepads/
That is a big list. Will take awhile to go through them all.

edited 3:22PM:
I went through that list and about 1/2 of the editors listed have broken links. Many of the remaining editors are not longer under development, have horrible ratings, require .NET, are slow launching, or are not portable. Out of remaining editors, three appeared worthwhile for further testing/investigation: Notepad3, XPad and Winvi32. I'm not sure if XPad and Winvi32 are truly portable (since settings are not in the app folder), but I will test them since they don't require installation and they seem to have a good feature set and seem fast. Notepad3 does save settings to an ini file in app folder.

I've seen the suggestion to use hex editors for large files before. That doesn't seem like it's a very user-friendly method, at least to me.

I tried PilotEdit Lite and Gloog a while back. Neither are portable like you mentioned. I had other issues with them as well, but can't recall the details now. There are many portable editors to choose from anyway.
Last edited by TP109 on Mon Feb 15, 2016 11:28 pm, edited 2 times in total.

TP109
Posts: 560
Joined: Sat Apr 08, 2006 7:12 pm
Location: Midwestern US

Re: Text Editor Performance Tests

#118 Post by TP109 » Fri Feb 12, 2016 5:11 pm

Tested WinVi32 v3.02, xPad v4.5 and NotePad3 bld 289.

Results uploaded to following link:
http://s000.tinyupload.com/index.php?fi ... 6253532816

TP109
Posts: 560
Joined: Sat Apr 08, 2006 7:12 pm
Location: Midwestern US

Re: Text Editor Performance Tests

#119 Post by TP109 » Fri Feb 19, 2016 10:28 am

Added installation size to previous results for ver 121:

http://s000.tinyupload.com/index.php?fi ... 4886964886

xor
Posts: 105
Joined: Sat Apr 18, 2015 11:02 pm

Re: Text Editor Performance Tests

#120 Post by xor » Sat Feb 20, 2016 3:01 am

What excel version is that .xls file? Couldn't open it in any Android office suite app, i.e. Excel, WPS, DocsToGo.

Post Reply