Justifying the removal because of its limitations isn’t fair. It was an advertised feature of the system. If I bought a PS3 to use as a Linux desktop, online multiplayer game machine, and SACD player based on their advertisement and then found I had to get rid of the first two to keep online gaming, on a $600 console, I’d be pretty upset.
At that time (before GPGPU become big for such purposes) scientists used PS3s to build cheap (much cheaper than IBM's server offerings containing a Cell processors) computing cluster for HPC. These PS3 computation clusters are were hard to program, but if you managed it, they were really fast (for their time).
Lets put aside the fact that Sony was probably unhappy that their loss leader console was being used in this manner(they did release the feature so maybe they could justify the PR bonus of supporting our troops or something).
All these researchers probably had no need to upgrade their firmware once they got set up since (correct me if im wrong) no firmware update actually improved the OtherOS functionality in any way so they lost nothing when a new firmware took it away. Only people really complaining about it were using it as a games (or piracy) machine and a Linux box to do...something?
> no firmware update actually improved the OtherOS functionality in any way so they lost nothing when a new firmware took it away.
At that time the respective scientists were quite concerned not to accidentally install this firmware update, and whether Sony would invent more draconian measures to disable consoles with old firmware.
This lead to a huge distrust in Sony, and accelerated the switch from the Cell to the emerging GPGPU technology for respective scientific calculation. Thus, this was some PR disaster for Sony.
Those scientists don't have good processes then. There are typically stringent procedures in regards to updating and handling software on laboratory computing equipment.
>This lead to a huge distrust in Sony, and accelerated the switch from the Cell to the emerging GPGPU technology for respective scientific calculation. Thus, this was some PR disaster for Sony.
Any proof of this? As far as I could tell, Cell was a dead end as a lot of the top tier benefits never really materialized irrespective of Sony shenanigans. Serious scientists used the server offerings from IBM. IBM eventually stopped supporting their server offerings for the most part. Why would they do this other than people stopped finding the architecture good for their needs?
