AI can interpret dyno graphs no worse that it does X-rays 
Loaded my graph with Specs into Claude.ai 3.5 is telling with this crap:
Here's a breakdown of what this chart is showing for your M20 stroker build:
Peak numbers
Torque curve shape
The torque curve is very linear and clean from about 2,000 rpm (65 ft-lb) up to peak at 4,730 rpm. No dips, no cam-induced lopiness in the pull — that's a sign of good tune and reasonably smooth VE growth, which tracks with a mild-ish 272° exhaust duration on the Schrick.
After peak, torque doesn't fall off hard — it holds a long, gentle plateau from ~4,700 rpm out to ~6,000 rpm (dropping only from 193 to roughly 180 ft-lb), which is unusual and actually a strength of this combo. Big-throat unmodified AMC head + ITBs + this cam are clearly keeping the head breathing well well past torque peak.
Power curve shape — the interesting part
Power isn't a single smooth hump — it shows two distinct humps:
That sag-then-recover pattern is a classic signature of intake resonance/runner tuning — with ITBs and a plenum, you get a runner length that's tuned to reinforce cylinder filling in one rpm band and mildly work against it in another. It's not a fault, it's just where your current runner length/plenum volume happens to null out slightly before coming back in on the second harmonic. Changing trumpet length (shorter velocity stacks or a different ITB horn) would shift where that dip sits.
The cliff at the top
Both curves fall off a cliff very abruptly after about 6,800–6,900 rpm — power drops from ~195 hp to near zero in a very short rpm window, and torque does the same. That's a much sharper drop-off than you'd expect from just running out of breath. A few likely causes, roughly in order of probability given your spec sheet:
Bottom line
Want me to walk through how to distinguish valve float from a rev-limiter cut in your logs, or talk through ITB runner length changes to fill in that mid-range dip?
Loaded my graph with Specs into Claude.ai 3.5 is telling with this crap:
Here's a breakdown of what this chart is showing for your M20 stroker build:
Peak numbers
- Max Power: 196.06 hp @ 5,910 rpm
- Max Torque: 193.51 ft-lb @ 4,730 rpm
- SAE correction factor 1.01 (test day was 90°F/45.79% humidity, so barely any correction needed — the numbers you see are close to true output)
Torque curve shape
The torque curve is very linear and clean from about 2,000 rpm (65 ft-lb) up to peak at 4,730 rpm. No dips, no cam-induced lopiness in the pull — that's a sign of good tune and reasonably smooth VE growth, which tracks with a mild-ish 272° exhaust duration on the Schrick.
After peak, torque doesn't fall off hard — it holds a long, gentle plateau from ~4,700 rpm out to ~6,000 rpm (dropping only from 193 to roughly 180 ft-lb), which is unusual and actually a strength of this combo. Big-throat unmodified AMC head + ITBs + this cam are clearly keeping the head breathing well well past torque peak.
Power curve shape — the interesting part
Power isn't a single smooth hump — it shows two distinct humps:
- A plateau around 185–188 hp from ~4,000–4,500 rpm
- A slight sag around 4,700–5,200 rpm
- A second rise to the true peak of 196 hp at 5,910 rpm, holding flat to about 6,200 rpm
That sag-then-recover pattern is a classic signature of intake resonance/runner tuning — with ITBs and a plenum, you get a runner length that's tuned to reinforce cylinder filling in one rpm band and mildly work against it in another. It's not a fault, it's just where your current runner length/plenum volume happens to null out slightly before coming back in on the second harmonic. Changing trumpet length (shorter velocity stacks or a different ITB horn) would shift where that dip sits.
The cliff at the top
Both curves fall off a cliff very abruptly after about 6,800–6,900 rpm — power drops from ~195 hp to near zero in a very short rpm window, and torque does the same. That's a much sharper drop-off than you'd expect from just running out of breath. A few likely causes, roughly in order of probability given your spec sheet:
- Valve float — you're running OEM valve springs with a fairly aggressive 284° intake cam. OEM springs are usually only rated to control the valvetrain to somewhere in the 6,000–6,500 rpm range depending on spring pressure and rocker mass. A sudden cliff like this (rather than a gradual power fade) is the textbook signature of the valve losing contact with the cam lobe at high rpm.
- Rev limiter/fuel or spark cut — worth checking your tune to confirm the limiter isn't set right at ~6,900–7,000, which would produce an artificially sharp cutoff that looks identical to valve float on the graph.
- Less likely but worth ruling out: ITB velocity/flow choking or exhaust backpressure at that rpm.
Bottom line
- ~196 whp / ~193.5 ft-lb from an M20 stroker is a strong result, especially with an unmodified head — power-per-liter suggests the ITBs, cam, and compression are working well together.
- The mid-range plateau and dip is intake tuning, not a problem — it's tunable if you want to fill it in.
- The sharp cliff past 6,800 rpm is the one thing I'd chase down before doing anything else, because if it's valve float on OEM springs, you're at risk of valvetrain damage (dropped valve, spring coil bind, or eventually a bent valve) if the car gets revved past that point regularly. I'd recommend a spring pressure/coil-bind check and consider upgraded valve springs matched to that cam before running it hard near redline again.
Want me to walk through how to distinguish valve float from a rev-limiter cut in your logs, or talk through ITB runner length changes to fill in that mid-range dip?
Comment