You tell'em Spine... that is a COMPLETELY UNACCEPTABLE result, no puttin' a nice face on it, no hand-waving it away, no sorry excuses will hide it, etc. 299 seconds to make the first box! For cryin' out loud... cheese and crackers...
I've seen similar results to the other two sets: 8 sec for the first box, 7 sec for the second box, 5 sec for the third and then 2 msec from then on) - hence the 20 seconds I've been stating.
And as the results indicate, it's the first 3 or so boxes, not just "the first 1 or 2". Note: this makes a significant difference in the startup time, and doubles the impact of the problem for the edit/compile/run cycle than if it really were only the first box. 8 seconds or 15 seconds or 299 seconds... it's ALL unacceptable for a simple program like this.
What if we were compiling something real? 10-15 mintues to compile (but of course it took less than a minute to compile the same program in DBC - they still haven't solved that little issue even in patch 4), then on top of that you have the stupid 299 second wait because DBS just didn't think it was worth fixing yet. ATI/Radeon users aren't worth all of the fixes they've been promised in patch 4 for the last four months - they'll have to wait until patch 5 or suffer every time they edit/compile/run one of their creations... not really a good impression to be giving the paying customers IMO.
It boils down to this:
1) The problem did not exist with patch 3, but now exists on many different computers (which all happen to be using a Radeon or ATI vid card), to varying degrees, in patch 4.
2) Evidently, they (DBS) knew about the problem before the release of patch 4.
3) They didn't document it on the patch 4 download page. There is no excuse for this and it's borderline deceitful IMO. Maybe they didn't have perfect information about the problem. Maybe the beta testers didn't push hard enough. But the sheer number of cases was probably enough to make them ask, "Hey, is there really something going on with patch 4 on all these Radeon systems?".
4) The best evidence of a solution we have is: "...we'll be looking into that for patch 5...". Aggravating. Plain and simple.
If someone doesn't make it crystal clear to DBS that the problem is unacceptable, then they will keep it on the back-burner and (maybe) fix it when they get around to it... I see the same situation at work with my under-staffed group of systems programmers - a simple case of "fire-fighting" or "reactive-programming" - fix the problems that are the most screamed about, while trying to press forward.
I personally want the added (and needless) 20 seconds of startup time that this bug causes to GO AWAY... for ME... NOW! But, I want all the fixes from patch 4 too - they've been a long time coming - and although I may be new to DB, well, maybe that just means I'm stupid enough to think that if I push for a fix, it will happen sooner than if I don't. Call me silly, call me whatever... you guys have been here longer than me, and you've probably had enough rants and tirades to last you. I'm 38 years old, and I'm on the receiving end of a lot of idiocy related to computers and software in general... it's amazing how these stupid little issues, a lack of attention to detail, can cloud the focus of the mind and ruin your whole day. Computers and software - it's all about focus and follow-through. Eventually, it all comes down to a choice made by one human being - do I follow through and solve the problem, or do I let it slip through the cracks another day. They're typically hard choices. I thought this weekend was going to be a milestone of achievement and grandeur for DB, a programming tool that is already an amazing tool that IS leaps and bounds ahead of everything else, taking a completely insane, over-engineered, over-designed, overly complex behemoth like DirectX, that follows no open standards except ones that M$ claims to make, and turning it into something simple and usable. And while patch 4 IS a good addition to that great achievement in a general sense, it falls flat on it's face in front of me in a matter of few minutes of actually trying it myself... because someone dropped the ball.
I bet, if the DBS team was forced to write DB code for a week on a Radeon system and had to wait 10-XXX seconds between every compile and a run test, they'd stop everything they were doing, take the time and fix the problem after the first hour or so.
