Is there a compare between MadExcept and Eurekalog

delphi package - automated exception handling
Post Reply
chuchiring
Posts: 21
Joined: Wed May 07, 2014 3:24 am

Is there a compare between MadExcept and Eurekalog

Post by chuchiring »

I need it moving to MadExcept.... thanks every one ! :D
madshi
Site Admin
Posts: 10754
Joined: Sun Mar 21, 2004 5:25 pm

Re: Is there a compare between MadExcept and Eurekalog

Post by madshi »

Hi there,

madExcept and Eurekalog overall have a quite comparable feature set. There are a few things madExcept does which Eurekalog does not. Probably there are also some things Eurekalog does which madExcept doesn't. But I'm not aware of any important feature Eurekalog would have that madExcept doesn't.

From my subjective point of view (and obviously I'm biased) madExcept has the following advantages over Eurekalog:

(1) madExcept is based on a tried and proven code base which was carefully extended for 64bit and Unicode support. In contrast to that large parts of Eurekalog 7 were rewritten, AFAIK, which have introduced all kinds of new problems in Eurekalog. At least several Eurekalog users have told me that. I've got a number of new madExcept users who were switching over from Eurekalog due to bugs in Eurekalog 7.

(2) madExcept doesn't slow down compiling/linking in any noticeable way. Eurekalog does slow down compiling/linking noticeably. At least that was the case the last time I checked.

(3) madExcept hooks over 400 APIs so that it can report all kinds of resource leaks to you (memory, kernel handles, user handles, GDI objects, etc etc). In contrast to that Eurekalog mostly only reports leaks in the Delphi memory manager.

(4) madExcept has an "instantly crash on buffer overrun" feature which allows you to find buffer overrun bugs in your code very easily. Eurekalog doesn't have this feature, AFAIK.

(5) madExcept has a low memory footprint and is very stable in multi-threaded environments. Eurekalog users have reported to me that Eurekalog 7's memory footprint is much larger, and that there are stability issues in multi-threaded applications.

(6) madExcept is free for non-commercial use, Eurekalog is not.

(7) madExcept has a very sophisticated stack tracing algorithm for 32bit processes which is able to produce very clean and exact stack traces. Eurekalog's algorithm is quite a bit simpler, resulting in more "noise" in the stack traces.

It's possible that some of the stability issues in Eurekalog 7 have been fixed in the meanwhile. I can't really say for sure. Also, as mentioned above, I'm quite biased, of course. So take my comment with a pinch of salt. My recommendation would be to try both products yourself and pick the one which works better for you. One more thing you could try is read through the support forums of both madExcept and Eurekalog and check the posts you see there. How many of them are bug reports? How many are just questions about how to interpret a bug report or things like that? Look at the ratio of bug reports compared to simple support/help requests. That should give you a general hint about the stability of each product. Also look at the support reply time and quality of answers, so you can get an impression about the support quality. Please take into account that madExcept probably has a bigger user base because it's free for non-commercial use.
chuchiring
Posts: 21
Joined: Wed May 07, 2014 3:24 am

Re: Is there a compare between MadExcept and Eurekalog

Post by chuchiring »

wow thank you man !!!
cubic
Posts: 10
Joined: Thu May 14, 2015 8:28 pm

Re: Is there a compare between MadExcept and Eurekalog

Post by cubic »

I liked EurekaLog 6. Last year I got a license for v7 hoping that it is waaaay much better than v6.
Turns out it is simply unstable. And the GUI is even more confusing that in v6. So, I returned to v6 (they have a good reason for which they still offer the few years old v6 in their download page).
About one week ago they released yet a new patch for v6. So, one more time I uninstalled v6, and tried the new release. Nothing changed.

The compilation time is HORRIBLE. No wonder since v7 is 5 (five) times bigger than v6 (when installed). I keep Eureka packages disabled all the time so I can compile my programs.
I activate Eureka packages ONLY when I build the final final release. But package management this takes precious time.

Eureka support never returned a helpful answer (they are actually quite rude). They blame Delphi XE7. They say Delphi XE7 is 'crude'. I don't know who to believe.

Today I sent them yet another two bug reports. They don't even bother. They said that they don't have enough info to reproduce the bug. For them, 'we cannot reproduce the bug' means 'the bug doesn't exist'.

They said, "Please send a compilable demo project to illustrate issue". I don't intend to send them my 54 PAS file project so they can identify the bug.

I got enough of this. I will try MadExcept.
madshi
Site Admin
Posts: 10754
Joined: Sun Mar 21, 2004 5:25 pm

Re: Is there a compare between MadExcept and Eurekalog

Post by madshi »

Thanks for your feedback, that's interesting to hear. I've got a couple of EurekaLog customers switching over to madExcept in the past couple of months. Seems many users aren't too happy with v7. So far the ex-EurekaLog users seem to be happy with madExcept.

FWIW, if a developer cannot reproduce a bug, it's really hard to fix. So I understand they asked for a test project. I sometimes do that, too. That doesn't mean that I'm not trying to help in other ways if a test project isn't possible, of course.

Anyway, if you run into any problems/questions, just let me know. Might be good to open a new thread then for such new topics. For now please feel free to simply install and try the non-commercial madExcept version. You can use it for a reasonably timed evaluation period, if you like.
cubic
Posts: 10
Joined: Thu May 14, 2015 8:28 pm

Re: Is there a compare between MadExcept and Eurekalog

Post by cubic »

Indeed too many people complaining about v7.
v7 is a dramatic step backwards compared to v6.
madshi wrote: FWIW, if a developer cannot reproduce a bug, it's really hard to fix. So I understand they asked for a test project. I sometimes do that, too. That doesn't mean that I'm not trying to help in other ways if a test project isn't possible, of course.
I do the same when a customer has a problem. But it is different. The client has so hand me one (input) file that doesn't contain critical information. But in my case I won't hand over a very large Delphi project.

Anyway, I believe (I am sure) they haven't tested their product enough for two reasons:
1. Too many customer complaining about v7
2. The problems/crashes/compilation time issues appear in most of my projects. So, it is not a one-time issue. It is easy to spot/encounter.

I expect to release a new version soon so I will need to activate Eureka one more time.
After this I will uninstall it an try MadExcept.
madshi
Site Admin
Posts: 10754
Joined: Sun Mar 21, 2004 5:25 pm

Re: Is there a compare between MadExcept and Eurekalog

Post by madshi »

Well, I know at least one thing you should be happy with: Compilation time. I've noticed that EurekaLog is usually *much* slower there compared to madExcept. Usually madExcept does its thing so fast you won't even notice it's enabled.
wpostma
Posts: 23
Joined: Fri Oct 12, 2012 4:09 pm

Re: Is there a compare between MadExcept and Eurekalog

Post by wpostma »

In my also-subjective opinion, I love MadExcept and do not like EurekaLog. Three reasons.

1. The Stability/Quality/Performance of MadExcept is all around better.

2. MadExcept bug report submission feature is better.

3. Support of MadExcept is better (Thanks Matthias).

Warren
madshi
Site Admin
Posts: 10754
Joined: Sun Mar 21, 2004 5:25 pm

Re: Is there a compare between MadExcept and Eurekalog

Post by madshi »

:D
Post Reply