By aspect — in detail
Tonality
Contested · 7 srcSources cluster three ways rather than two: a balanced-monitor reading, a warm-and-slightly-muffled reading, and a dry-and-lean reading. The measured curve is close to a target through the bass and lower mids but deviates hard on either side of 3 kHz, which is enough for careful listeners to land in different places honestly.
Measured
On MRSallee's KEMAR (KB006x) traces, read against that database's KEMAR-DF 10 dB-tilt target and normalised 200–1000 Hz, the curve is within about half a decibel from 50–200 Hz and averages +0.2 dB across 300–2000 Hz, then falls roughly 9 dB below target across 2–4 kHz before a +4.5 dB peak near 8.4 kHz and a 10–16 kHz region some 6 dB down.
Where it splits
reads as balanced, monitor-neutral40%
“Really quite balanced. Somewhat emphasized in the middle, incredibly detailed, but overall very balanced.”
Eugen Wegmann · PCWorld
reads warm, and a little muffled up top19%
“they would be especially suitable if you are looking for comfortable, powerful and warm-sounding headphones for use over several hours”
Carsten Kaiser · headphonecheck.com
reads dry, lean and fatiguing41%
“The resulting timbre is very fast, dry and stale.”
GREQ · Head-Fi
Sources split on whether the low end is complete. Reviewers quoting the 10 Hz spec describe effortless, deep bass; listeners comparing it to other planars describe a competent mid-bass with no sub-bass slam. The trace supports the second reading and explains the first.
Measured
The KEMAR traces put 50–200 Hz within about 0.4 dB of target, then roll off below: −2.7 dB averaged over 20–50 Hz and −6.2 dB at 20 Hz.
⚠ vs. listeners — The 'deepest bass' claims track Fostex's 10 Hz specification rather than a measured shelf; the trace shows the driver still moving that low but several decibels down, which is exactly what a listener expecting planar slam would report as missing.
Where it splits
mid-bass is fine, sub-bass is missing58%
“Bass lacks the sub-bass punch and texture you might expect from larger planars.”
GREQ · Head-Fi
The most polarising axis. One group hears a natural, well-integrated midrange; a second hears it shoved forward around 1 kHz; a third hears female vocals hollowed out. The measured step between those two regions is large enough that all three descriptions fit the same curve.
Measured
300–2000 Hz averages +0.2 dB against the KEMAR-DF target and peaks at +2.6 dB near 1 kHz, while 2–4 kHz averages −9.2 dB (−10.0 dB at 4 kHz) — an eleven-decibel step down across the region where female vocals sit.
⚠ vs. listeners — 'Too much midrange' and 'hollow female vocals' are the same measurement described from either side of the step: in absolute terms the lower mids are on target, but the missing pinna gain above them makes 1 kHz read as forward and 3 kHz as absent. Swapping to deeper pads lifts 2–4 kHz by about 3 dB in the same measurement set, which is why pad-rollers report the effect fading.
Where it splits
glaringly forward through the lower mids28%
“there is too much energy in the midrange between 300 and 2000hz that makes all music sound too glaringly mid-range forward”
GREQ · Head-Fi
hollowed out through the female-vocal range24%
“Anything in the female vocal range (or instruments that sit in that general range) are probably the worst I've ever heard that range, on any over-ear headphone above the $100 price point, and that's a huge detriment to ship with.”
klardotsh · Head-Fi
Three readings coexist: a peak in the 8–10 kHz region that treble-sensitive listeners find piercing, a restrained top octave that a test lab heard as muffled, and fine resolution that a studio reviewer rated above an AKG K702. The trace shows all three are present at once.
Measured
7–9 kHz averages +1.6 dB against the KEMAR-DF target with a +4.5 dB peak near 8.4 kHz, while 10–16 kHz sits about 6.5 dB below it.
⚠ vs. listeners — The disagreement is not about the graph. A narrow peak below a rolled-off top octave is heard as glare by listeners sensitive to 8–10 kHz and as darkness by listeners who miss the air above it — and different rigs read that region very differently, which is why the lab and the enthusiast measurements generate opposite adjectives.
Where it splits
peaky and fatiguing around 8–10 kHz47%
“The peak at 8KHz however, is unfortunately quite present”
GREQ · Head-Fi
restrained, short of air on top31%
“The treble reproduction of the Fostex T50RPmk4 was warm and restrained and the super high-frequency range was also rather restrained. This made the overall sound a little muffled.”
Carsten Kaiser · headphonecheck.com
detailed and finely resolved22%
“In direct comparison, it falls behind the Fostex at least in terms of treble reproduction.”
Harald Wittig · 7Review
Imaging
Strong consensus · 5 srcThe one thing nobody argues about. Editorial, community and critical sources all single out localisation and separation, and it is the reason Fostex ended up marketing a gaming-branded version.
“the imaging ability of this headphone which, at least for my head or personal HRTF, is one of the very best I’ve ever experienced”
GREQ · Head-Fi
“the headphones offer incredibly good spatial audio, i.e. they can show very well where something is located in the game”
Eugen Wegmann · PCWorld
“Here's where Mk4 start pulling its distance with its predecessor, the layering is just better.”
ed7090 · Head-Fi
Soundstage
Contested · 5 srcMostly reported as closed-in on stock pads and noticeably wider on deeper aftermarket ones — the split tracks pad depth, not listener taste. One studio reviewer using the headphone as supplied still called the spatial picture very good.
Measured
The pad dependence is visible in the measurement set: the ZMF hybrid-perforated trace lifts 2–4 kHz from −9.2 dB to −6.0 dB against target, the region that most affects perceived distance and openness.
Where it splits
closed-in stock; opens up with deeper pads79%
“Soundstage is much wider and more spacious, imaging has not taken a big hit in accuracy and timbre has evened out a lot to something much closer to realistic.”
GREQ · Head-Fi
Broadly positive — the new driver is repeatedly described as fast and resolving, including by listeners who dislike its tuning. The lab that measured it is the lone dissenter, calling resolution ordinary and blaming the busy treble.
“Mk4's better, it just resolves better and has a better transparency.”
ed7090 · Head-Fi
“the T50RP mk4 is a really good magnetostat. It always delivers detailed, quite naturalistic images of the audio material.”
Harald Wittig · 7Review
“Due to its rather ‘busy’ sound in the highs, the signal resolution was rather ordinary.”
Carsten Kaiser · headphonecheck.com
Transient speed is the driver's calling card and is praised almost universally; the one qualification is that attacks are described as broad rather than sharply pointed, and perforated pads reportedly soften them further.
“As befits magnetostatic headphones, the T50RPmk4 also has strengths in the dynamics. The dynamic image differentiation ranged from good to satisfactory, with percussive transients being reproduced as broad rather than pointed.”
Carsten Kaiser · headphonecheck.com
“This new driver excels technically and even has decent timbre for a planar, not normally something I find planars do well.”
klardotsh · Head-Fi
“As expected bass and dynamics took a hit, and remained dry”
bisa3d · Head-Fi
The sharpest practical disagreement. A test lab and two studio reviewers found the clamp-to-weight balance close to ideal; long-session listeners report a headband hotspot within half an hour, heat under the pleather pads, and stock pads too shallow for comfort. Almost everyone who swaps the pads says the problem goes away.
Measured
headphonecheck measured 326 g without the cable (369 g with) and a clamping pressure of 676.5 g averaged across its large and small head jigs — mid-pack for a studio planar, which is consistent with the split being about headband contact area rather than raw force.
Where it splits
clamp and weight are well judged out of the box57%
“The combination of pressure from the headband and the weight of the headphones was almost perfect.”
Carsten Kaiser · headphonecheck.com
headband hotspot and hot, shallow stock pads43%
“the headband padding continuing to bring a hotspot on the top of my head after 30 minutes or so”
GREQ · Head-Fi
Isolation
Moderate · 3 srcSemi-open by design, and sources treat it as a deliberate trade rather than a failing: enough damping to track with a microphone in the room, nowhere near enough to shut out a commute.
“they let through more ambient noise than closed headphones, but less than open headphones”
Eugen Wegmann · PCWorld
“These headphones have better isolation to the outside, so that microphone recordings—keyword: crosstalk—are feasible.”
Harald Wittig · 7Review
Measured
headphonecheck measures exterior noise damping alongside frequency response; Fostex's own closed sibling, the T50RPmk4CL, is described by the maker as the same internal housing with the vents sealed.
Nobody calls it luxurious and nobody calls it fragile. It is high-strength plastic with metal adjustment rails, a two-metre OFC cable and no case; the notable upgrade this generation is a symmetrical shell that takes the cable on either side, and long-time RP owners rate its assembly quality above earlier runs.
“The workmanship is solid and offers no cause for complaint.”
Harald Wittig · 7Review
“the earpiece shells are made of admittedly somewhat cheap-looking plastic”
Eugen Wegmann · PCWorld
“Luckily, T50RPmk4 has higher quality control like T60RPmk3 and T50RP-50th.”
Philimon · Head-Fi
“The T50RP, however, uses Tip for L+, Ring-1 for R+ (which is the standard) and Ring-2 (which would be sleeve in any 3.5mm TRS jack) for L- and Sleeve for R-.”
solderdude · Audio Science Review
Measured
The single-entry socket is a 4-pole 3.5 mm TRRS wired for balanced use, which is why ordinary dual-entry replacement cables misbehave — an owner who tried two of them heard only the left channel.
Genuinely split, and the split is about what you are buying. Judged as a finished headphone it draws satisfied verdicts on price-performance; judged against the cheaper mk3 that mods just as well, or against rivals with published measurements, it looks expensive for a tool that most owners will modify anyway.
Where it splits
steep for something that still needs modifying45%
“300 euros feels like a steep price for a tool with limited use cases that absolutely requires modification, and even more spending for audio-pleasure.”
GREQ · Head-Fi
