My 6TT reads 4mph high at indicated 75 mph, with 305/30 and 245/40.
My other 6TT (blue) reads 4mph high at indicated 75 mph with 295/35 & 225/40 which are stock sizes. With 315/35 and 245/40 it's spot on.
My cayenne reads spot on at all speeds on stock tire sizes.
My VW's read 3-5 mph high at 75 indicated. No digital display in the VW's, just read the needles.

 

I don’t usually go off topic but who remembers these? I wonder about the accuracy given its manufacturer. Although is doesn’t have any electrical components.

 

08 Toyota + 3 mph
02 C2 also +3 mph
71 MGB GT who knows? Speedo bounces around

 

I'm 3-5 MPH high depending on the speed. Faster = farther off.

 

Yeah my 99 C2 is about 3mph. It looks like brand new tires vs. worn tire could be with 1-3mph.

 

I've had a strange thing happen with my car. I've had it for over 13 years and it always read 3 MPH faster than the actual speed. I know that is by design for the above mentioned reasons. However, since about 2 years ago it displays the actual MPH as confirmed by GPS and radar.

 

We should be careful here about looking for precision. There are a lot of error sources.

Brand new 295/30R18 on my C4S have a circumference of 1992 mm. If I were to run off all 9/32 of tread, the outer circumference is now 1947 mm, or about 2.2% less. So if my Speedo was accurate with new tires at 70mph, the old tires will now read 71.6, or 2.2% faster. (So if it was 3mph fast before, it's now 4.6mph fast.)
My understanding from the article linked above was that sometimes even different manufacturers with the same tire sizes, can differ a little in outer circumference.
What width rim is the tire on? Is it stretched? That can change outer circumference.

GPS isn't perfect either. With positional error of about 16ft, and up to 32ft between two points used to calculate velocity, that's 70*(32/5280)=0.42mph error. Error can be worse if you are in a hilly or mountainous area, or in buildings were there may be multipath reflections.
And phones usually only update at about 1Hz (although I think they could do 10Hz if programmed for it?), so for any measurements, make sure your speed is constant on level terrain, and wait one second to check your GPS speed.

I may be wrong on some of my assumptions above specifically, but my overall point is that while we can say that our Porsche speedometer seems about 3mph fast, anything more specific is kind of a waste of time without rigorous data collection and error analysis.
This is exactly why ECE-R39 makes manufacturers read fast. That buffer accounts for unknown error on individual cars. They feel that the error determined by that equation is sufficient so that no car should ever read slower than it is actually going.
So if they rigorously calculated a potential error of about 3mph, I don't think we're going to guess our way to be much more precise than that.

 

I take that back, checked my new Cayman today, it's dead-on accurate! Surprised me, but my 03 Carrera is off by 2-3mph.

 

My '03 Targa with Tiptronic and new 285/3018 rear tires reads 2-4 MPH higher, depending on speed. This seems consistent with the information reported above.

 

Mine is 4 MPH fast in the 60 to 80 MPH range.

 

I found with most cars, that it's usually pretty accurate at city speeds, and the error margin increases with speed. If I remember correctly, on my old BMW E90 it could be programmed to either show the speed precisely or with the "error" (and at ~260km/h on the speedo, GPS displayed something about 245km/h - not a huge deal imo).

 

my 996 C4S - high by 2 mph
my 958 GTS - high by 3 mph
my 970 GTS - high by 3 mph (also noticed that the cruise is 1 mph than the set speed...set cruise at 79, speedo reads 80?!)

 

So you're telling me when my speedo said 159 I was only going 156??

 

Skip was right

 

bury the hatchet??