The preload before, would load things that may never be needed.
The new graphics engine is more efficient. That is one reason why older computers are running the new version better than the old version while the new version looks better as well.
How is it possible for older computers to run the new version bettter when you have to dial down the in game graphics settings to get decent frame rates when it wasn't necessary with the 2.13 version? I seem to remember a massive outcry from people using older systems with older video cards when 2.14 was first released, and if that were the case then my P4 with the 7300GT should be able to get as good if not better in game performance with the same settings (512 textures, bumpmap and detailed terrain on, all other off).
Or are you referring to older systems with upgraded video cards? I could understand that since a P4 processor can handle the load, or should be able to and a simple video card upgrade could put enough hardware power into it to overcome any limitations.
If you are getting stutters regardless of how the "Detailed Terrain" option is set, then it is something else not releated to rendering. It has to be.
I'm not totally sure either way Skuzzy, that's why I'm asking. If I had access to the code I could find out for myself...but not going there...so I'm just tossing out the ideas based on my observations and knowledge of 3d rendering. Could it be a bottleneck problem with not enough pixel pipelines or maybe 128bit vs 256bit interface? There was one patch (5? or 6?) that the stuttering went away...and I didn't see anything in the release notes that showed a change that would account for it.
It's not going to be something big or necessarily obvious...could just be a small item that was thought to be "unnecessary". All I know is 2.13 ran totally awesome with 512 textures and all other options active on my 7300GT AGP but since 2.14 my much better 9500GT PCI-E 2.0 on a better hardware platform is having issues that never appeared in the previous version or on my older system...and I'm not the only one.
For me this isn't a huge deal for me...just an annoyance that I'm fairly certain can be resolved from the programming end.
5400RPM SATA drives are slow as snails. A 25% increase in the rotational latency and a reduction in the areal density, is going to slow things down. It cannot help but do so. It is painfully obvious if you put a 7200RPM drive next to the same computer with a 5400RPM drive.
Yeah that's true, I was more referring to data path flow....which the SATA standard is the same across the board regardless of drive speed. Seek and read/write times are going to be better on a 7200 rpm drive, but once you get to the point of data flow it would be the same.