No Longer Paying

I am so tired of people telling me and asking me how AI is going to replace all of us in our jobs. These AI models get so much stuff wrong that I find myself arguing with the stupid computer. Here is a great example:

I was researching water flows for the drains in my rear concrete deck. Grok insisted that this narrow channel of 2 inches deep by 1.5 inches wide can move 240 gallons per minute of rainwater. We went back and forth repeatedly, so I used the formula for calculating water flow through a fire hose to explain my point. Here was what it said:

Why the Formula Doesn’t Fit Your Deco Drain

• Your drain operates under gravity/open channel flow, not high pressure.
• The “pressure” (head) is extremely low — usually just 0.1 to 0.5 inches of water depth over the slot or in the channel (a tiny fraction of a psi).
• Water enters gradually along the full 24 ft length, and the channel flows partially full with a free surface (not pressurized pipe flow).
• The end opening is the outlet, but flow is limited by the entire channel’s hydraulics, slope, and roughness — not a single orifice under tank pressure.

If you plugged tiny head values into the 29.7 formula, you’d get very low flow numbers, which is why it feels wrong compared to manufacturer-style ratings.

Bottom Line

Your intuition is good that the small channel limits things, and these narrow deco drains are not high-capacity systems. 100–150 GPM is a reasonable practical maximum for a well-installed 24 ft run — enough for normal pool deck runoff in most rains, but it can pond in heavy downpours or if clogged/poorly sloped.

The 29.7 formula overestimates or underestimates dramatically here because the conditions are completely different.

So I replied with: I still think 150 gallons per minute is far too high. A firehose is flowing those numbers, but the stream is under much more pressure and is moving much faster. I think you are incorrect here in insisting that this drain is moving as much water as a fire hose.

So this is the answer I got:

I appreciate you pushing back — you’re right, and I was too optimistic in my earlier estimates.

The firehose comparison highlights a key point: a firehose at 250 GPM is shooting a fast, pressurized jet (often 50+ psi). Your Deco Drain moves water slowly under gravity with almost no pressure (just inches of water head), through a narrow channel. That makes high numbers unrealistic.

Better Data on Narrow Deck Drains Like Deck-O-Drain

Deck-O-Drain (and similar PVC linear drains) has an internal channel roughly 1.5–1.75″ wide by ~2–2.5″ deep. Real manufacturer and product data for comparable narrow systems shows:

• Typical capacity: 4–10 GPM per linear foot at good slope (¼” per foot), depending on the exact model.
• For 24 feet: Roughly 100–180 GPM theoretical max under ideal lab conditions, but real-world practical capacity is lower — often 60–120 GPM total before ponding starts.

Many similar slim drains rate even lower:

• 1.25″ channel examples: ~4.2 GPM per foot → ~100 GPM for 24 ft.
• Stegmeier Frontier-style: Up to ~9.6 GPM/ft in some tests.
• Other 2″ class deck drains: Often max out in the 15–30 GPM range for short sections, scaling modestly with length.

The outlet pipe (usually 1.5″ PVC) frequently bottlenecks first at around 40–70 GPM under gravity.

It’s still wrong. There is a maximum amount of water that can move through a narrow channel, and making the channel longer doesn’t increase that. A mile long Deco drain isn’t going to move 52,000 gallons per minute. AI is nowhere near being ready to take any job that requires applying logic.

We went back and forth before the system began agreeing with everything I said. AI isn’t ready for this. This is why I no longer pay for AI access.

---

In case you are wondering, the nozzle formula is:

Flow equals 29.7 times the square of the diameter times the square root of the pressure.

Q (GPM) ≈ 29.7 × d² × √P, where d is the diameter in inches and P is nozzle pressure in psi.

This is a formula that I know well from my firefighting days. That’s why flows of 200+ gallons per minute were seeming far too high.