intric8 wrote:Fascinating
alternative opinion by internetsux (in response to Bulletdust's post) on Reddit:
The main issues with WHDLoad are that you rely on patches being made for the games you want, there are like only 3 people making patches actively (it’s mostly StingRay) who can’t keep up with bugs and shit because it’s overwhelming, and the patches do NOT run well on all hardware. Blitter Waits are a shit show making lots of games have slowdowns on anything below an 040. Look at all the EAB requests for patches to be redone because of slow downs. And if you have like an A500 with an ACA the games will tun worse than via floppies.
It is convenient and nice but the way it works means it excludes a lot of hardware to run it on optimally, even a lot of A1200s.
I love it, don’t get me wrong (I bought two licenses back in the day), but it’s not really perfect or ideal.
JOTD’s CD32Load seems to run much better but it’s still very experimental.
I have to say Intric8, I've never experienced an issue with blitter waits and I'm running an 030.
As stated, when it comes to 16bit OCS/ECS Amiga's and faster 030 and above processors, there is a problem with wait states when certain applications hit the chipset directly that is purely a result of interfacing a 32 bit processor with a 16 bit bus, the issue has nothing to do with WHDLoad at all - This is an issue myself and another friend both took up with Jens when the ACA500 was not long released and was confirmed by Jens himself. This issue is present on the original .ADF as well as the original floppy running outside WHDLoad and is no worse running inside WHDLoad - The second you insert the accelerator in a 32 bit Amiga and run the same software there is absolutely no issue as you aren't inserting wait states to interface a 32 bit bus to a 16 bit bus.
This is 100% the reason why I don't run accelerated 16 bit OCS/ECS Amiga's anymore - Quite often they actually run 'slower' once accelerated.
I get everything from WHDownload and I've never had an issue with fragmentation of slaves, everything runs perfectly and I've run WHDload on both an accelerated A500 as well as an accelerated A1200 using the exact same OS install.
Respectfully, I don't find WHDLoad feels experimental at all.