Teal Towers vs Tensor Farms

Choose your station

Welcome to the Silicon Graphics Web Navigation Console. Below you can configure the browsing arsenal that powered the artists, physicists, and dreamers of the ’90s—and pit those teal towers against today’s tensor-farm fantasies. Select a workstation, align an IRIX release, and see how the price/performance story evolved from SGI glam to H100 grind. Warning: by the time you scroll to the price banner you will want an SGI, an H100 rack, or both.

Timeline of Machines & Browsers

1991 – 1993

Indigo / Indy (IRIX 4.x → 5.2)

CPUs: MIPS R3000/R4000   |   RAM: 24–128 MB
Browsers: NCSA Mosaic (Motif), early Arena builds
Static HTML 1.0, inline GIFs, no frames. Perfect for rendering teapots while you read a CERN press release.

1994 – 1996

Indigo² / High Impact (IRIX 5.3 → 6.2)

R4400 at a blistering 250 MHz. Netscape Navigator 2.x arrives: tables, frames, JavaScript 1.0, and the ability to crash spectacularly on malformed `` tags.

1996 – 1997

Indy (late) & O2 (IRIX 6.3)

Unified Memory Architecture meets Netscape 3.01 Gold. JavaScript 1.1, animated GIFs, and livestreaming QuickTime if you sacrificed enough RAM. Great for coding VRML scenes at 3 a.m.

1997 – 1999

Octane / Octane2 Launch (IRIX 6.4 → 6.5.8)

Dual-channel graphics, IMPACT rails, Netscape 4.05–4.78. CSS 1, DOM Level 0, and the occasional browser freeze whenever DHTML gets ambitious. Still, you can monitor your render farm and browse Slashdot simultaneously.

2000 – 2003

Octane2 (late) / Fuel

Community ports of Mozilla 1.0/1.2 via Nekoware. CSS 2 partial support, better PNG handling, but startup times counted in coffee breaks. Comparable to what a G3 Mac could pull off with Mozilla 1.0.

2003 – 2006

Tezro / Prism

R16k or Itanium horsepower. Firefox 1.5 ports exist, but most SGI veterans switch to x86/Linux for modern browsing and keep the SGI purely for Maya, Flame, or massive OpenGL sims.

2017 – 2020

GPU Renaissance

NVIDIA RTX 2080 Ti, RTX 3090. Chrome, Firefox Quantum, and the rise of CUDA. Hobbyists build “AI rigs” that could vaporize an Indigo PSU.

2022 – 2025

Tensor Farm Era

H100, DGX Spark, eight-way HGX trays. PyTorch/LLAMA/Claude on demand. You can serve GPT-4 sized models from your garage if your power company approves.

Configuration Modules

Browser Arsenal

  • NCSA Mosaic (1992) – IRIX 4.x
  • Netscape 2.x (1995) – HTML 3.2
  • Netscape 4.78 (1999) – CSS 1, JS 1.2
  • Mozilla 1.0 (2002) – community port
  • Links/Lynx – when in doubt, go text

Hardware Flights

  • Indigo R4000 – Entry imaging suite
  • Indy Video – Live digitizing + web
  • O2 Cam – Unified Memory for VRML
  • Octane MXE – Dual-channel geometry
  • Tezro – Quad pipe, onyx DNA

Off-the-shelf Upgrades

  • IRIX 6.5 overlay CDs (18+)
  • Nekoware browser builds (Mozilla/Firefox)
  • Gopher gateways for time travel
  • SSH tunnels to your Retro AI Proxy

Tensor-Farm Add-ons

  • RTX 4090 — GPU sledgehammer for proxy bursts
  • RTX 5090 — Next-gen raytraced hype machine
  • DGX Spark — $3,999 command deck in a box
  • H100 PCIe (80 GB) — When you need datacenter swagger
SGI CLASSICS — INDIGO: $12,995 (≈ $26k today)   |   OCTANE2: $74,995 (≈ $130k today)   |   TEZRO: $110,995 (≈ $170k today) MODERN AI — H100 PCIe (80GB): ≈ $30,000   |   HGX 8× H100: ≈ $250,000   |   DGX Spark: $3,999   |   DUAL RTX 4090: ≈ $3,000
Finance? Sure. Just mortgage the render farm.

Compatibility Notes

No Internet Explorer. Microsoft never ported IE to IRIX. Netscape dominated, with late-era relief from community Mozilla/Firefox builds. Expect IE5-grade content to render better on your G3 Mac than on an SGI; the Octane2 with Mozilla 1.0 will match a Mac running the same vintage build, but nothing beyond that.

JavaScript Reality. Netscape 4′s JS 1.2 is roughly IE4-level. Dynamic HTML will break. The Retro AI Proxy solves this by rendering psuedo-live UIs that target IE5 features but degrade to iframe polling—exactly the kind of trick an SGI could use to stay useful in 2025.

Use Cases. Monitor SGI news feeds, order visual effects middleware, render check your latest Maya scene, and yes—run your Retro AI Proxy in a teal window, because you can.

Reality Check: 1999 vs 2025

Wonder what that teal tower would cost in today’s AI arms race? Adjust the dials below and compare the sticker shock.

SGI Indigo R4000

$12,995 (1991) / ≈ $26,000 today
R4000 100 MHz, 64 MB RAM, GR2-XZ graphics
AI perf: ≈ 0.00005 tokens/sec — suitable for ASCII palm trees

SGI Octane2 Dual-Pipe

$74,995 (1999) / ≈ $130,000 today
R12000 400 MHz, MXE graphics, 2 GB RAM
AI perf: ≈ 0.0003 GPT-4 tokens/sec (mostly renders teapots)

SGI Tezro Quad-Pipe

$110,995 (2004) / ≈ $170,000 today
4× R16000 700 MHz, V12 graphics, IRIX 6.5.22
AI perf: ≈ 0.0007 tokens/sec if you bribe IRIX with coffee

NVIDIA H100 PCIe (80 GB)

≈ $30,000 (2025 street)
80 GB HBM3, 3.5 PFLOPS FP8
AI perf: ≈ 2.5M GPT-style tokens/min

HGX 8× H100 Baseboard

≈ $250,000 barebones
8× H100 NVLink, 28 PFLOPS FP8
AI perf: ≈ 20M tokens/min (and 20k BTU/hr)

NVIDIA DGX Spark

$3,999 (desktop kit, 2025)
1× Spark AI SoC, 32 GB HBM3e, 400 W PSU
AI perf: ≈ 400k tokens/min — think “personal GPT lab”

Dual GeForce GTX 5090

≈ $3,800 (MSRP×2)
48 GB combined GDDR7, 1 kW draw
AI perf: ≈ 1.2M tokens/min with mixed precision

Dual GeForce RTX 4090

≈ $3,000 (MSRP×2)
48 GB combined GDDR6X, 900 W
AI perf: ≈ 750k tokens/min, still cheaper than therapy

CONCLUSION: SGI DREAMS ALWAYS DEMANDED DEEP POCKETS. TODAY’S AI RIGS STILL WANT YOUR VC, YOUR GARAGE, AND PERHAPS YOUR FIRSTBORN—BUT THE POWER PER DOLLAR IS FINALLY ON YOUR SIDE.