[p4] CVS vs. Perforce - Stephen Vance - com.perforce.perforce-user

7 messages in com.perforce.perforce-user[p4] CVS vs. Perforce

From	Sent On	Attachments
Todd Short	10 May 2002 11:23
Hoff, Todd	10 May 2002 11:35
Stephen Ng	10 May 2002 11:41
Stephen Vance	10 May 2002 11:51
Stephen Ng	10 May 2002 16:43
Barrie Slaymaker	13 May 2002 11:20
John Marshall	16 May 2002 07:43

Subject:	[p4] CVS vs. Perforce
From:	Stephen Vance (ste...@vance.com)
Date:	05/10/2002 11:51:20 AM
List:	com.perforce.perforce-user

At 02:23 PM 5/10/2002 -0400, Todd Short wrote:

Question:

Perforce supports 3-way merges (base, your changes, my changes), which is something we like. We want to stay with Perforce, but we're merging with another group that uses CVS and doesn't see the benefits of Perforce.

* Does CVS support 3-way merges? automatically?

Yes, but it doesn't keep track of already merged (or not) chunks.

* Has anyone used WinCVS (www.wincvs.org), how does it compare to P4Win? How about tkcvs (http://www.twobarleycorns.net/tkcvs.html)? We don't have the time to setup a CVS repository to really test these out.

I've used WinCVS and found it to be usable, but frequently obscure. I've also found a number of the macro add-ins to just plain not work, even with Perl and Python installed and configured properly.

* I've seen a lot of comparisons of CVS vs. Perforce, obviously the key things are atomic changelists and better merge/branch support. Is there anyway I can present this information to make it overwhelmingly obvious that Perforce is the way to go?

Stage a competition. Define a process flow problem in which a development branch is created, changes are made and merged back in a couple of rounds, and a release is staged and let loose on QA and release branches.

See if their best CVS user can keep up or even complete the task without administrative permissions.