I haven't gotten it in the months that I have been making my map. Now all of a sudden the error is occuring B1 to B1a. Here’s what’s happening: Based on some of the feedback I got on my map (thanks for the feedback by the way), I sealed the top level of the two canyon areas with nodraws, displacements and area_portals - this has really upped the fps in the small part of the canyon by 30-80 fps and about 10-20 in the larger part (what can I say - it's pretty wide open in the large part – hard to optimize it). The only other thing I've done is to re-do the elevator in the water - instead of having it go from the river to the 1st ledge it now goes up to all three levels making it easier for players to get back into the fight depending on what level the fight is. I have made the elevator structure 1 func_detail and added in 2 more path tracks. Now all of a sudden, the dam thing is crashing all the time – i.e. load map, pick a pryo, do a flame burst and get the crash. I’ve been looking around and have followed some of the advice here: http://forums.tf2maps.net/showthread.php?t=426&highlight=crashes I have deleted at least 20 entities and converted 80 dynamic props into statics and have even gone to the extent of taking all my current func_details making them into world details – doing a select all and then making them into 1 func_detail. When checking the ents via the map>show information – I am within the limits posted in the above link too. Any help in this would be greatly appreciated because I can’t understand why the error wouldn’t happen in the first version and then suddenly appear in the second. Also, is there a faster way to come up with this number “entdata” without having to do a full compile to see what my load is? The issue might also be in the compile itself – for some reason it looks like it’s trying (ok doing) vrad twice in hammer – the first seems to complete and then the second one is just hanging there and I have to do a force quit of it – this error only happens when I do a full compile, running bsp and rad alone/together are ok, it seems to happen when I add vis into the mix (reg vis not fast).