[breck]

I expect to pay for this with high probability. I don't think I'm in the first target batch as I'm giddy in M1 land now, but I do work on so many different machines and love the idea of a persistent environment in the cloud. I also expect to want to do genomics in my browser at some point, and thus envision a need for 100x+ more powerful browser tabs.

[choxi]

What would you be doing that would require 100x+ more powerful tabs? I'd imagine most process-intensive work is already being done server-side or in a desktop app, not the frontend of a browser app.

[breck]

Someday I want to run a whole world simulation. Think "The Sims" except the whole world. 8 billion agents, say a million bytes per person, so 8PB of RAM. While the sim is running I want to copy and paste the URL in a new tab and change a few params to compare the results. I want things to be instant.

Today I want to visualize 100,000 rows across 1,000 dimensions in 10 different tabs.

Between Today and Someday there are endless things I want to do.

[shaklee3]

But that doesn't really have anything to do with the browser, which is what the op was asking.

[breck]

"I want to copy and paste the URL in a new tab and change a few params to compare the results."

[discreteevent]

You are not explaining your architecture. The parent (and me as it happens) assume that when you paste the URL and change params then that URL is sent from your browser to a server. The server runs the simulatíon based on the params in the URL it received and returns the results to the browser. With that architecture you would need a lot of resources for the server but not for the browser. What architecture are you thinking of?

[breck]

Server is just a dumb nginx server sending HTML and Javascript. No dynamic routes. Everything happens clientside (main thread and/or web workers, local storage for persistence).

Same architecture as: https://v20.ohayo.computer/?filename=discovery-of-elements.o...