You have room for one cardio machine, and someone lives under the floor. The second fact decides more of this than the first, and it usually arrives as an adjective: bikes are low impact, treadmills are loud. An adjective cannot tell you whether to buy the treadmill. Numbers can. So this guide puts them on the three things that settle it in a flat: what a running stride drives into the slab, how much floor a folded treadmill really hands back, and what has to be true before a resistance level means anything at all.
Start with who is under you
If an occupied unit sits below you, buy the bike. I’d treat that as settled rather than as a preference, and the rest of this guide is the reason why.
Three exceptions are real. A top floor, or a floor above an empty or commercial unit, removes the constraint and puts you back to choosing by training goal. A dedicated room with a door changes what you are willing to live with. And if you are weather-blocked or building-blocked with no stairs worth using, a treadmill can earn its space, as long as you buy it knowing what folding does and does not give you.
What “high impact” weighs
Impact is measurable, and it has been measured. A force-plate study of 100 recreational runners recorded peak vertical ground reaction force at 2.37 times body weight at 9 km/h, 2.51 at 11 km/h and 2.62 at 13 km/h. Those are treadmill figures, from healthy recreational runners, and the useful part is the slope: the same machine is a different problem for the flat below at 8 km/h than at 13.
Calculation
An 80 kg runner at 11 km/h: 80 kg x 2.51 = about 200 kg of peak force, once per footfall. At a cadence of 160 steps a minute, a 30-minute run delivers roughly 4,800 of them. The 80 kg body mass and the 160 spm cadence are assumed inputs, not study findings. Substitute your own and the arithmetic holds.
A bike puts none of that into the floor. Pedalling is a seated, continuous load, and the machine’s mass never leaves the ground between strokes. Which is why the two are not really competing on noise. They are competing on whether the floor gets struck at all.
Noise is not a decibel question
The number on the box measures the wrong half of the problem. Airborne noise, the motor hum and the belt, moves through air and leaks past gaps, doors and thin walls. Impact behaves differently: contact with the floor sends vibration into the structure, and it travels through joists and slabs before radiating back out as sound in the room underneath. Acoustical Surfaces states the practical consequence directly, that a wall with a high STC rating will not stop a footfall problem any more than an isolated floor will stop voices.
So a treadmill’s published noise rating answers the airborne question and is silent on the one your neighbour is complaining about. A dense mat under the deck is still worth having, for the floor and for the machine, but it reduces something rather than removing it, and no product page can tell you what is left in the ceiling below. The material side of that is in home gym flooring, and the wider set of quiet-hours tactics is in quiet apartment workouts.
Folded is not the same as smaller
Three machines whose manufacturer pages published a full dimension set, read on 2026-08-29. Named to represent their categories, not as recommendations.
| Machine | In use (L x W) | Folded | Floor area, in use to folded | Height, in use to folded | Weight | Max user |
|---|---|---|---|---|---|---|
| Schwinn IC4 (bike) | 54.6 x 30.7 in | does not fold | 1.08 m2 | 132 cm | 113.6 lb | 330 lb |
| Sole F63 (treadmill) | 75 x 34 in | 42 x 34 in | 1.65 to 0.92 m2 | 132 to 180 cm | 187 lb | 325 lb |
| Horizon T101 (treadmill) | 70.5 x 33.5 in | 46 x 34 in | 1.52 to 1.01 m2 | 150 to 155 cm | 180 lb | 300 lb |
Two numbers in that table do the work, and neither is the one people shop on.
The first is folded area. Folding the F63 releases 44 percent of its floor and the T101 releases 34 percent, which reads like the storage problem solved. Then compare what is left against the machine that never folds.
Calculation
Folded, the Sole F63 stands on 0.92 m2 and the Horizon T101 on 1.01 m2. The Schwinn IC4 stands on 1.08 m2 and stays there. So a folded treadmill occupies 7 to 15 percent less floor than a bike occupies permanently. Areas are length times width from the manufacturer dimensions in the table: 42 x 34 in is 1.07 x 0.86 m, or 0.92 m2, and so on for the rest.
The second number is height, and it is the one that changes the decision. Both treadmills get taller when folded. The F63 goes from 132 cm to 180 cm. Folded, it becomes a shoulder-high panel standing in the room: nothing sits on top of it, it will not go under a bed, and you walk around it every time you cross the floor. The bike, at the same 132 cm, at least reads as furniture and will hold a towel.
I’d stop treating “foldable” as a storage feature on machines this size. It is a way to reclaim a walkway during the day, which is worth something, but it does not hand the room back.
Weight decides whether any of that folding actually happens. The F63 is 73 lb heavier than the IC4, and 187 lb raised and lowered before and after every session is a chore you will end up negotiating with. If you are buying a treadmill for a flat, that 187 lb is the number I’d shop on, well ahead of the 3.0 HP motor that gets top billing. Static mass is not the interesting figure here, though: 85 kg spread across 1.65 m2 is an ordinary furniture load, and what a floor notices is the 200 kg spike arriving 4,800 times.
Illustrative example
Suppose a 3.2 by 2.6 m living room with a sofa along one long wall taking 0.9 m of depth, leaving a clear zone of about 3.2 by 1.7 m. A Sole F63 at 1.91 by 0.86 m fits inside that on paper, with 0.84 m spare across the width. Then add what the machine needs in order to be used: clearance behind the deck to step off, and a route past it to the door. That 0.84 m is the walkway, and the treadmill has just taken it. The room dimensions here are invented. The point is that “it fits” and “the room still works” are two different measurements, and only one of them is on the spec page.
What a resistance level has to mean
European standards draw a line that the marketing does not. Under EN 957-1/5, a Klasse A ergometer has to be adjustable to at least 250 watts and display that power to within 5 percent up to 50 W and 10 percent above it. DIN 32932 sets its own Klasse A at the same 250 watts with a 10 percent display tolerance, and equipment in that class is cleared for medical and therapeutic use. Below it the standard is blunt: a Klasse B home trainer may not show power in watts at all, and expresses resistance in stages instead. Klasse B needs a flywheel of at least 5 kg; Klasse C sits under that.
The consequence is simple enough. A level is an internal scale, not a unit. It is not comparable between two machines, and on a bike that is not rated to display power it cannot be. The Schwinn IC4 lists 100 magnetic resistance levels and publishes neither a flywheel mass nor a power figure, which is ordinary for the category, and is exactly why the level count tells you nothing about how hard level 60 is.
So if a bike does not publish a watt figure, do not compare its resistance levels against another machine’s, and do not lean on them to track progress across a change of equipment. Levels are meaningful only against themselves, on one machine. That is a shopping rule I apply, not something any manufacturer states.
Whether that matters depends on how you train. If you ride to a plan and want a number that survives a new bike, watts are the thing to pay for. If you ride to feel and to the clock, a level dial covers it, and I would not spend extra on calibrated power just to have it.
Matching a bike effort to a run
Calories per minute scale with body mass, which makes them a poor bridge between two machines. METs do not, so they are the honest way across. From the 2024 Adult Compendium of Physical Activities:
- Stationary cycling: 101-125 W is 6.8 MET, 126-150 W is 8.0 MET, 151-199 W is 10.3 MET.
- Running: 6 mph (a 10-minute mile) is 9.3 MET, 7 mph (an 8.5-minute mile) is 11.0 MET.
Read across, a hard but sustainable bike effort of 151 to 199 watts sits at 10.3 MET, between a 10-minute mile and an 8.5-minute mile. That is the trade the bike is offering: a comparable metabolic cost, delivered without the 200 kg spike. The WHO asks adults for at least 150 to 300 minutes of moderate activity a week, and either machine clears that easily. Which one you are still using in March is the part no spec sheet decides.
A narrower note about knees
This comparison usually turns into a question about joints, so it is worth being precise about what the data above can and cannot support.
The force figures describe a load, not an injury. A higher peak ground reaction force is not the same thing as harm, and the study cited here measured force on a treadmill; it did not follow anyone’s knees over time. What can be said is mechanical and narrow: cycling is a seated movement in which the pedals and saddle carry the body’s mass, so the load through the leg per revolution is far below a footfall of two and a half times body weight. That is a difference in loading, and it is a plausible reason a bike stays available on days when running is uncomfortable.
It is not a treatment, and it is not a claim that a bike protects a joint or that a treadmill damages one. If you have knee pain, an existing injury, ongoing pain, pregnancy-related considerations or limited mobility, that belongs in a conversation with a clinician rather than in a purchase decision.
Which machine, by flat
| Your situation | Buy | Why |
|---|---|---|
| Occupied unit directly below | Bike | About 200 kg per footfall goes into the structure, and no mat removes it |
| Mid-floor, neighbours above and below | Bike | The worst case for structure-borne impact, and folding does not buy the room back |
| Top floor, or above an empty or commercial unit | Either | Choose by training goal; the constraint that decides the rest is gone |
| Dedicated room with a door | Either | Space and noise stop competing, so buy for the training |
| Shift worker or a sleeping child below | Bike | Timing is the whole problem, and impact does not care what hour it is |
| No safe outdoor walking, no stairs worth using | Treadmill | A real need. The T101 is the walker’s machine; the F63’s 20 x 60 in deck and 12 mph range are what running actually asks for |
If someone lives under you, buy the bike, and buy it for the reason that outlasts every other argument in this comparison: it is the machine you can switch on at 6 a.m. without thinking about it first. Everything else here is a preference. That one is a constraint.
If nobody is below you, or the room is genuinely yours, the treadmill is a real option, and I’d buy it on the deck and the speed you will actually use rather than on the fold. Walkers are covered by the Horizon T101. Runners want the Sole F63’s 20 by 60 inch deck and its 12 mph range, and will live with its 187 lb. Either way, measure the walkway before the machine: floor area is the easy part, and the route past it is what people get wrong.
The strength half of a home setup is in the small home gym setup guide, the smaller-machine version of this decision is in walking pad vs treadmill, and the wider apartment context is in staying active in a small apartment.
Sources
- Peak vertical ground reaction force in running, force-plate study of 100 recreational runners (PMC, 2.37-2.62 x body weight at 9-13 km/h)
- Cross-Heimtrainer - EN 957-1/5 and DIN 32932 ergometer/home-trainer classes (watt control, display accuracy, flywheel mass)
- 2024 Adult Compendium of Physical Activities - stationary cycling MET values
- 2024 Adult Compendium of Physical Activities - running MET values
- WHO guidelines on physical activity and sedentary behaviour (World Health Organization, 2020)
- Airborne vs Impact Noise (Acoustical Surfaces)
- Schwinn IC4 Indoor Cycling Bike (manufacturer specifications, read 2026-08-29)
- Sole F63 Treadmill (manufacturer specifications, read 2026-08-29)
- Horizon T101 GO Series Treadmill (manufacturer specifications, read 2026-08-29)
