Emulator detection

[HAL6128]

Just a quick question:
I need a emulator detection for an application. I've read in the Wiki the possibility to read out port &FEFE: if the returning value is different to &FF (= real CPC) then it's an emulator. But some emulators (like WinApe, Caprice) return the same value like a real CPC.
Is there a better way to detect if an application works (or not) inside an emulator?

Thanks.
HAL6128

[andycadley]

I'd say pretty much any "reliable" emulator detection is a bug.

[Zoe Robinson]

Surely the point of any emulator is to emulate the original hardware, so it would return any value the actual CPC would return?

[arnoldemu]

Surely the point of any emulator is to emulate the original hardware, so it would return any value the actual CPC would return?

exactly!

[arnoldemu]

Just a quick question:
I need a emulator detection for an application. I've read in the Wiki the possibility to read out port &FEFE: if the returning value is different to &FF (= real CPC) then it's an emulator. But some emulators (like WinApe, Caprice) return the same value like a real CPC.
Is there a better way to detect if an application works (or not) inside an emulator?

Thanks.
HAL6128

My question:  What are you doing that doesn't work on emulators?

If you would share the code with me I can make sure my emulator works with your code and that I can tell other people how to make their emulators work better too

[AMSDOS]

Sprites Alive sets a New Frame Flyback Event Block (&BCD7) if you used that (along with the New Frame Flyback Event Block) then do a OUT &7F00,&C4, the next bit of code following that could simply state emulator.

[TotO]

A way to detect emulators is to check the FDC count.

[arnoldemu]

A way to detect emulators is to check the FDC count.

fdc count?

fdc timings?

[TotO]

The RPM is around 300 on a real CPC, but never return the same value.
Just knowing that allow to said witch emulator is running...

Best, on a 3" floppy the RPM decrease on each track from 0 to 39 (40, 41) because the belt...
On a 3"1/2 floppy, the value don't change so much. (on HxC too)

Last year, I have made a little program showing that, if you remember my screenshot showing all emulators detected?

[Bryce]

Best, on a 3" floppy the RPM decrease on each track from 0 to 39 (40, 41) because the belt...

How does the belt and track number have any relation to the RPM??

Bryce.

[Lazy Dude]

That's a belter of a question.

I'd like to say that all the emulators I have tried in 464 mode always have a virtual disc drive attached, there is no option to 'disconnect'  this and have a standard 464.

[TotO]

How does the belt and track number have any relation to the RPM.

The belt allow the floppy to turn, but it is flexible and vagueness.
Depending if you are inside or outside the disc, this gap is more or less important and you never got the same RPM speed on a same track.

[TotO]

That's a belter of a question.

I'd like to say that all the emulators I have tried in 464 mode always have a virtual disc drive attached, there is no option to 'disconnect'  this and have a standard 464.

And a virtual PAL attached too... The memory is not properly handled for a 464.
As I know, only SugarBox handle all those things.

[Bryce]

The belt allow the floppy to turn, but it is flexible and vagueness.
Depending if you are inside or outside the disc, this gap is more or less important and you never got the same RPM speed on a same track.

Ah, now I understand what you mean = The drift is more noticeable on the outer tracks. I thought you were claiming that the speed was actually changing depending on what track you were reading.

Bryce.

[TotO]

Sorry, I'm tired and I have difficulty to explain things today. 

[gerald]

On a 3" drive the speed is slightly changing according to the track, but this due to the head/disk contact resistance and motor torque.
Inner track (0) will spin faster than outer track (39). But the difference is in the 5 rmp range.
This should not be that visible on drive that use a direct drive like 3 1/2 drive.

[arnoldemu]

The RPM is around 300 on a real CPC, but never return the same value.
Just knowing that allow to said witch emulator is running...

Best, on a 3" floppy the RPM decrease on each track from 0 to 39 (40, 41) because the belt...
On a 3"1/2 floppy, the value don't change so much. (on HxC too)

Last year, I have made a little program showing that, if you remember my screenshot showing all emulators detected?

I forgot about it

I understand now

[arnoldemu]

And a virtual PAL attached too... The memory is not properly handled for a 464.
As I know, only SugarBox handle all those things.

So does Arnold WIP

[TotO]

@gerald: Are you sure about the motor torque? Because, if you goes up and down into the tracks number, the RPM value shift for the same track.
On my CPC, the tests shown that is more 2RPM than 5RPM between track 0 and track 39.

[Targhan]

In the Midline Process, from what I remember, I detected an emulator by getting the ID of the sector under the head, made a track seek, read the next sector, and checked that there was a difference of 2 in the ID. It was pretty reliable, but I guess emulators have improved since, so it may not be enough.

[dragon]

Nothing

[TFM]

On a 3" drive the speed is slightly changing according to the track, but this due to the head/disk contact resistance and motor torque.
Inner track (0) will spin faster than outer track (39). But the difference is in the 5 rmp range.
This should not be that visible on drive that use a direct drive like 3 1/2 drive.

Isn't track 0 outside and track 39 inside (close to the center)?

[gerald]

Isn't track 0 outside and track 39 inside (close to the center)?

Yes 

[remax]

So does Arnold WIP

[TFM]

Oh! You got a picture of my cat!