About performance

I have some questions, please help me to answer:

1.Can Gameshell enough run psp & ps1 games by emulator (e.g.,RetroPie,PPSSPP)
(Please official test.Thanks)

2.Can I replace Gameshell’s motherboard with Raspberry Pi 3?

3.About Gameshell’s performance and GPU, compared with Raspberry Pi 3, which is better?

Thank you.

  1. Not sure really and until we have the thing in hand.

  2. Well no, not the same form factor, not the same connection.

  3. It is like comparing an ATI vs an nVidia. The Broadcom GPU on the Pi is not that recent, the Mali is not the most powerful (and you need drivers for both which are not part of the Linux Kernel)

From an “able to handle it” perspective, you can compare the specs of the PSP with the GameShell.

Emulation does not work this way.

The PSP and PS1 are using a MIPS CPU which need to be emulated to run on another system. PS1 games are emulated on the PSP and PS Vita, so with a properly made emulator, that would probably run (not even thinking about other problems**)
For the PSP, as it is more powerful than the PS1, can’t say for sure, maybe, I never used a PSP emulator so not sure, but well a PC made emulator will probably not work well. (if not properly optimized)

** like no analog stick, original PS1 game with no need for analog may work with the light key module for the 4 shoulder buttons. PSP games that use the analog stick would be in the exact same situation, and there are way more than the PS1

1 Like

I was more trying to make the point that if you’re wanting to emulate a game that expects 4GB of RAM and a modern processor, emulating it on the the GameShell is not standing a good chance. The GameShell has about 8x more RAM than the PSP and the PSP measures it’s processor in MHz, so from purely a hardware standpoint, it at least has a chance of emulating these games. This is why I said:

Oh ok sure.

Well on a pure hardware point of view, the 6502 at ~1Mhz is super slow compared to a modern 3Ghz PC, hence, making an accurate emulator of a NES is not that easy without eating a lot of CPU.

1 Like

I don’t have the hardware at my hand now, but the team has shared some CPU and IO benchmark testing result that might be useful in reference.
According to this Raspberry Pi benchmark, the CPU speed is above Pi2 level.

I/O: Sequential writes is around Pi3 level, well that probably just show the performance of a decent SD card, anyway.


what bogomips this processor ?

1 Like

For a second I thought that was a typo or a made up term. BogoMips, I guess you learn something new every day.

BogoMips, as it name say is bogus. This is a value that the kernel generate internally to try to calibrate an internal timer. This is everything but a way to compare CPU performances.

By the way, I didn’t know but it seems to be useless on ARM cpu:

In 2012, ARM contributed a new udelay implementation allowing the system timer built into many ARMv7 CPUs to be used instead of a busy-wait loop. This implementation was released in Version 3.6 of the Linux kernel.[8] Timer-based delays are more robust on systems that use frequency scaling to dynamically adjust the processor’s speed at runtime, as loops_per_jiffies values may not necessarily scale linearly. Also, since the timer frequency is known in advance, no calibration is needed at boot time.

One side effect of this change is that the BogoMIPS value will reflect the timer frequency, not the CPU’s core frequency. Typically the timer frequency is much lower than the processor’s maximum frequency, and some users may be surprised to see an unusually low BogoMIPS value when comparing against systems that use traditional busy-wait loops.


tell me how speed cpu is in game shell?


1 Like

What is difference in new processor?

bogomips in dmesg