Hey Tim,
I know we already spoke directly and went over all of this, but I feel its a good idea to clear these up here as well for the benefit of everyone reading.
It's very difficult to compare results from two different dynos, let alone two different cars on two different dynos. There are many factors at play that can alter power as well as spool readings. The first major variable is gearing/loading. Our EQT Turbo results were done on a Golf R with DSG on our Dynocom Chassis Dyno. The VTT results were done on a GTI (not sure if manual or DSG) on a Dynojet. The gear ratios on the GTI are significantly taller than the Golf R. As most people are aware, a taller gear ratio applies more load to the engine which results in significantly faster spool. For power pulls, our dyno is always ran in inertia only to keep results consistent. I'm not sure if VTT applied any additional eddy brake loading during their testing. Another major factor is ambient temperature. Colder temperatures significantly reduce spool time as well. During the testing of our turbo, our dyno room recorded ~90F ambient temps. I don't know the weather for VTT's testing, but they are in a significantly cooler area in general. Finally, the supporting mods can have a big impact on spool as well. Our test car was running a catted downpipe and stock axle-back exhaust. Those do create a bit of a restriction and can slow down spool. I don't know the details on the VTT test car, so I can't compare there. Any one of those factors alone can easily explain a difference of spool of 200-300RPM. The point here is that you really can't put any weight on small variances in dyno charts when there are this many variables involved.
This is one of the bigger misconceptions in the post. Our dyno is a Dynocom chassis dyno with 24" rollers. We used to run the dyno with a .85 correction factor to make it read like the older low reading mustang dynos. This means that we were artificially making the dyno read 15% lower than every other Dynocom dyno in the world. We did this when we first got the dyno about 9 years ago because we were tuning nothing but Subaru's and our main competitors at the time were running the lower reading Mustang dyno. Those numbers became the standard for the Subaru community back then, so we decided to calibration to that. Over the years, more and more companies started using calibrations that read more like Dynojets (roughly 15% higher). Mustang Dyno even released a new version of their software that made most new Mustang dynos read just like Dynojets. Over the years we also expanded to other platforms as well as product sales where low reading mustang numbers did not make sense to customers. After years of explaining why our dyno reads so low, we decided in 2018 that we would switch to the native Dynocom calibration with no artificial correction factor. Our dyno now reads just like every other Dynocom dyno in the world which is very close to most Dynojets, Dynapaks, and Mustang dynos with newer software. At the end of the day, the dyno is really just a tool and the important thing is how the car performs in the real world. I personally don't care if our dyno reads high or low as I know our cars perform very well on the road and on track. But of course people like to compare results, so we try to be as consistent as possible.
As far as your specific car and results, I am convinced you had an issue when you ran it on VTT's dyno (or VTT has the lowest reading dynojet in the country). I have had customers go back/forth between our dyno with its current calibration and several Dynojets and the results are always within a few %. My guess is that you had a bad tank of gas and the car was pulling a lot of timing, or your IS38 was on its way out. I know your turbo ended up making noise and going out fairly soon after those dyno pulls, so that seems like a likely cause of the discrepancy. The other issue is that VTT doesn't seem to have any other IS38 baseline numbers. If you look around at Dynojet results for IS38 GTI's and Golf R's, your results on VTT's Dynojet are very out of the norm. The main point here is that you can't use a single data point like this to compare how two different dynos read, especially when that data point seems out of line with the norm for a particular dyno and setup.
I hope this clears some things up. As always I'm happy to answer any other questions anyone may have.
Thanks
-- Ed