You descend on your first dive of the day to 60 feet. Your dive computer displays 51 minutes of no-decompression limit time before you must surface. Your buddy’s computer, made by a different manufacturer, shows 56 minutes. You’re breathing the same air at the same depth, in the same water. So why the five-minute difference?
The answer lies in decompression algorithms. An algorithm is the mathematical formula dive computers use that factors in real-time measurements of depth, time at depth, water temperature, gas mix, and cylinder pressure to calculate how long you can safely remain underwater without mandatory decompression stops. Different manufacturers interpret the science of nitrogen absorption differently, apply different safety margins, and weight different risk factors with their own proprietary conservatism settings. This variation is not a flaw—it’s fundamental to how decompression theory works.
This article reveals how different algorithms work, why they produce different no-decompression limit (NDL) values for identical dives, and what that means for your dive planning and bottom time.
The Four Core Decompression Algorithms
Most dive computers today use one of four main decompression models, each with a different approach to predicting how your body absorbs and releases nitrogen. Understanding these models helps you understand why your computer behaves the way it does.
The Bühlmann ZHL-16 algorithm uses 16 tissue compartments, developed by Swiss physician Dr. Albert Bühlmann in the 1960s. Each compartment absorbs and releases nitrogen at different rates. Fast compartments (with 4-minute half-times, like blood and lungs) reach saturation quickly but also off-gas quickly. Slow compartments (with half-times up to 635 minutes, like bone) load slowly but release slowly as well. Bühlmann calculates safe ascent rates based on controlling compartment, which is the tissue compartment closest to its maximum allowed nitrogen pressure. This compartment determines your NDL.
The Reduced Gradient Bubble Model (RGBM) focuses on microbubbles, developed by Bruce Wienke in the late 1990s. Instead of preventing all bubble formation (as Bühlmann assumes), RGBM assumes tiny bubble nuclei already exist throughout your body. During ascent, these seeds can grow if pressure reduces too quickly. RGBM controls bubble growth by adjusting decompression schedules and requiring deeper initial stops. Suunto dive computers use proprietary RGBM variants in their recreational models.
The Varying Permeability Model (VPM), developed by David Yount and refined by Erik Baker, also focuses on bubble formation but applies Boyle’s Law to model bubble expansion during ascent. VPM-B (the Baker variant) often produces deeper initial decompression stops and shorter shallow-water decompression time compared to Bühlmann profiles. Shearwater offers VPM-B as an optional upgrade for technical diving.
The Diving Science and Technology (DSAT) algorithm developed for PADI, created by Dr. Raymond Rogers, modified US Navy research for recreational diving. DSAT allows longer bottom times on multi-level dives than square-profile tables because it credits you for time spent at shallower depths. Oceanic dive computers offer DSAT as their primary recreational algorithm.
How Algorithm Conservatism Changes Your Bottom Time
Real NDL Differences at Recreational Depths
When Suunto tested their RGBM algorithm at 60 feet, their computer calculated 51 minutes of no-decompression time on a first dive, compared to 56 minutes on the PADI Recreational Dive Planner table. At 100 feet, Suunto showed 17 minutes while the RDP showed 20 minutes. This 10 to 15 percent conservatism difference is consistent across recreational depths.
Bühlmann-based computers like Shearwater tend to fall between RGBM and DSAT in terms of conservatism, depending on gradient factor settings. With default gradient factors of 40/85 (medium conservatism), Shearwater computers typically show bottom times slightly longer than Suunto but shorter than pure DSAT implementations.
Naval Experimental Dive Unit research validating commercial algorithms confirmed that Bühlmann ZHL-16C, VPM-B, EMC-20H, and Suunto-RGBM produce markedly different results. The study found that at depths shallower than 30 meters, computers tended to be more conservative than standard tables, but between 30 and 50 meters, some computers were less conservative than tables.
Adjusting Conservatism With Gradient Factors
Bühlmann-based computers allow gradient factor adjustment to control conservatism using two numbers: GF Low controls the depth where your first decompression stop begins, and GF High affects how long you spend at shallow depths. The range spans from 10% to 99%.
A high conservatism setting like 35/75 shortens NDL time and requires deeper initial stops. A low conservatism setting like 45/95 extends your NDL by reducing mandatory decompression overhead. This flexibility allows experienced divers to fine-tune their risk profile, but it also means two divers using identical Shearwater computers might receive completely different NDLs if they’re using different gradient factor settings.
Suunto’s RGBM includes mandatory safety stop requirements of 3 minutes at 10 feet for any dive deeper than 30 feet, making the algorithm inherently more conservative for recreational dives regardless of user settings. Oceanic’s dual-algorithm computers let you switch between Pelagic DSAT (more liberal for warm-water dives) and Pelagic Z+ (more conservative for repetitive cold-water diving).
Why Multi-Level Dives Expose Algorithm Differences
Real-world dives are rarely square profiles where you descend to maximum depth, stay there, then ascend. Most divers follow the contour of the bottom or move to shallower water, spending part of their time deeper and part shallower.
Dive computers track actual depth continuously for credit, crediting you for every minute at shallower depths. This means multi-level dives on computers often provide 20 to 40 percent more bottom time than tables predict. Dive tables assume your entire dive happens at your maximum depth, so a 30-minute dive to 100 feet (with 5 minutes at 60 feet) gets penalized as if you spent all 30 minutes at 100 feet.
But that multi-level advantage varies by algorithm. DSAT-based computers designed for recreational multi-level diving maximize this benefit. VPM-B computers, designed for technical dives with deep stops, distribute their decompression time differently than Bühlmann models, sometimes requiring significantly more shallow decompression time.
Algorithm Penalties on Repetitive Dives
After your first dive, nitrogen remains in your body. On your second dive, all algorithms penalize you—you get less NDL time because you start with residual nitrogen. But the magnitude of that penalty varies by algorithm.
RGBM algorithms like Suunto apply heavier penalties on repetitive dives, especially if you make multiple dives in a day. Experienced technical divers express frustration with this, believing Suunto over-penalizes even with substantial surface intervals. Bühlmann-based algorithms are less reactive to repetitive dive history, requiring more diver knowledge to avoid diving too aggressively across a multi-dive day. This is why understanding your algorithm matters—it affects how much time you lose on dives two, three, and four of your day.
Selecting an Algorithm That Matches Your Diving
Matching Algorithm to Your Dive Profile
If you do single recreational dives in warm water with long surface intervals, DSAT-based computers reward you with longer NDL times. You’ll notice a significant bottom-time advantage compared to RGBM computers. Oceanic computers with dual-algorithm capability let you select DSAT for these dives.
If you do multiple dives per day or dive in cold water where your body might off-gas more slowly, RGBM or conservative Bühlmann settings provide safety margin that reduces your DCS risk across the dive series. You’ll spend less time at depth but exit each dive with greater physiological safety.
If you plan to move into technical diving or want maximum flexibility, Bühlmann-based computers with adjustable gradient factors give you control. You can be liberal (45/95) on simple shallow dives or conservative (30/70) on complex deep dives, all on the same computer. This flexibility comes with responsibility—you must understand gradient factors and when to apply different settings.
Why Some Algorithms Remain Proprietary
Suunto, despite publishing some RGBM details, keeps their exact algorithm parameters proprietary. This means you can’t predict precisely how your Suunto will behave on an unfamiliar dive profile. You must trust the computer or dive with conservatism settings until you know it well. Shearwater publishes their Bühlmann implementation entirely, letting users understand (and modify) exactly how the algorithm works.
This transparency difference matters. If your buddy’s Suunto shows 17 minutes NDL but your Shearwater shows 20 minutes, and you don’t understand either algorithm deeply, you might blame the computers. In reality, you’re seeing the algorithms’ different philosophies in action—different methods of interpreting the same nitrogen physics.
Five Questions to Ask Your Dive Computer
- Does your computer’s manual specify which base algorithm it uses—Bühlmann, RGBM, VPM, or DSAT? (All manufacturers publish algorithm specifications)
- Can you adjust conservatism settings, and if so, does the manual explain what those settings do numerically? (Conservatism factors range from zero to fifty percent)
- Does your computer apply mandatory safety stops or are they optional? (Suunto requires three minute safety stops)
- On a trial dive to your favorite depth, does your planning function show NDL times matching your buddy’s computer? If not, do you know which algorithm each uses? (Different computers often show varying NDL values)
- Have you tested your computer in multi-level dives against a table prediction to see how much extra bottom time it gives you? (Computers often provide longer times than tables)
Scoring: If you checked 3 or fewer items, you’re likely diving with limited understanding of your computer’s logic. Consider reading your manual’s algorithm section and doing a practice planning session before your next dive trip. If you checked 4+ items, you’re diving responsibly and understand the assumptions your computer makes.
What Your Dive Computer Cannot Know
Algorithms Are Mathematical Approximations, Not Certainties
Most recreational divers believe that staying within their computer’s NDL guarantees they won’t get decompression sickness. This assumption is wrong. Hyperbaric physicians treating DCS patients report misconceptions that many patients insist they couldn’t possibly be bent because they stayed within their computer’s limits. But here’s the reality: decompression sickness is probabilistic, not deterministic.
Your dive computer cannot measure the actual nitrogen in your tissues. It models theoretical nitrogen loading using compartment assumptions. A diver with a patent foramen ovale (PFO—a hole in the heart wall), poor fitness, or certain medications will off-gas more slowly or experience bubble formation differently than the algorithm predicts. Cold water, physical exertion, and dehydration all increase DCS risk beyond what your computer can calculate.
All algorithms include safety margins to account for uncertainty, but not all margins are equal. RGBM’s bubble-focused approach includes margin for bubble dynamics. Bühlmann’s dissolved-gas approach includes margin via M-values and gradient factors. Neither perfectly predicts your individual physiology.
The Haldanean Bubble Problem Affects Standard Tables
Haldanean algorithms descend from 1908 research foundations. Haldane’s genius was grouping the body’s tissues into compartments and measuring gas loading rates. But his model assumed gas was always released at the same rate it was absorbed, and it ignored tiny “silent bubbles” that form during ascent and slow off-gassing.
Unmodified Haldanean algorithms tend to be deceptively liberal because they predict you can surface faster than is actually safe because they don’t account for bubble-induced off-gassing delays. This is why RGBM and VPM algorithms were developed: to address the bubble reality that pure dissolved-gas models miss. Modern computer algorithms compensate for this, but the underlying assumption difference persists.
Why Experienced Divers Plan Conservative
Research shows that proximity to NDL increases DCS probability significantly when diving multiple dives in a day. A diver consistently touching the NDL across five dives in a day exposed themselves to significantly more cumulative decompression stress than a diver staying 20 percent away from NDL.
This insight drove a conservatism strategy: if your computer says you have 45 minutes of NDL, plan your dive for 35 minutes. The algorithm gives you safety margin but add your own. Experienced diving doctors recommend treating your computer’s NDL as an absolute maximum, not a target.
How to Use Algorithm Knowledge in Real Dive Plans
Managing Computers With Different Algorithms
You and your buddy are geared up for a dive to 80 feet. Your Shearwater computer shows 28 minutes NDL. Your buddy’s Suunto shows 22 minutes. Before descent, you need a decision: do you both ascend at 22 minutes, or does the Shearwater diver get extra time?
The safest approach is following the most conservative computer. This isn’t because one algorithm is “wrong”—it’s because decompression sickness is probabilistic. If you don’t understand why your computers disagree, you shouldn’t split the difference.
Better: before each dive trip, have one diver (or both) bring their computer’s planning function and agree on expected NDLs at your planned depths using the same gas mix and first-dive assumptions. If they differ, talk through why. If one diver is doing repetitive dives and the other is on their first dive, their computers may legitimately show different NDLs. Understanding the reason prevents frustration and unsafe decisions.
Adjusting for Conditions Algorithms Can’t Measure
Standard algorithms assume you’re a healthy, hydrated, rested diver at sea level in normal water temperature. Real dives rarely meet all those conditions.
At high altitude above 300 meters elevation, atmospheric pressure is lower, reducing your oxygen and nitrogen partial pressures. Your dive computer must be set to the correct altitude range, or its calculations will be dangerously wrong. Some computers allow adjustment in 500-meter elevation bands.
In cold water (below 20 degrees Celsius), your blood circulation slows and off-gassing rates decrease. Algorithms can’t know your body’s response to cold, so conservative divers reduce NDL time by 10 to 15 percent in cold water. If you’re tired, dehydrated, or diving after days of previous dives, additional conservatism is justified.
How Gradient Factors Reveal Your Decompression Stress
Shearwater computers display GF99 showing your saturation percentage, which shows your current saturation as a percentage of the absolute limit. At the surface before diving, GF99 is 0%. As you descend and load nitrogen, GF99 rises. During ascent, a properly programmed computer keeps GF99 below your set limit (typically 85% on the surface).
In an emergency where you must ascend immediately, watching your GF99 tells you how close to the danger zone you are. If GF99 is at 60 percent you have safety margin. If GF99 is at 85%, you’re right at your algorithm’s limit. This single number connects all the compartment calculations into one understandable value—your decompression stress level right now.
Other manufacturers don’t display GF99, so you lose this transparency. Understanding your algorithm deeply means knowing what numbers your computer is tracking behind the scenes.
Your Algorithm Choice Affects Every Dive You Make
The five-minute difference between your Suunto and your buddy’s Shearwater at 60 feet is not a manufacturing error. It’s the result of two different mathematical philosophies trying to solve the same problem: how to keep you alive while letting you enjoy your dive.
Bühlmann-based algorithms treat dissolved nitrogen as primary risk and use theoretical tissue saturation limits to determine safety margins. RGBM algorithms also track dissolved nitrogen but add bubble nucleation theory, applying extra conservatism to control bubble formation. DSAT prioritizes multi-level dive credit. VPM-B emphasizes deep stops to eliminate micro-bubbles early.
None of these approaches is universally “best.” Each makes trade-offs: more NDL time versus more safety margin, or depth-focused decompression versus shallow-focused decompression. Your job as a diver is to understand which trade-offs your algorithm makes, then dive responsibly within those assumptions.
Before your next dive, spend five minutes with your computer’s manual or planning function. Understand which algorithm it uses. See how it calculates NDL at your planned depth. If you dive with a buddy using a different computer, do a 30-second planning comparison before you descend. That simple step understanding your algorithm protects your safety. That might be the most important decision you make for your bottom time and safety.