MPC delay and reverb: 23 entries, six with a real name

The short version

This site's own insert vs send/return effects article already counted all six categories in Akai's Effects & Parameters appendix. Delay/Reverb won, 23 individual effects, ahead of Modulation's 22 and every other category:

"98 individual effects split across 6 categories… Delay/Reverb 23, Dynamics 14, EQ/Filter 17, Harmonic 19, Modulation 22, Vocal 3, total 98."

Only six of those 23 get Akai's full treatment, a named description paragraph and a dedicated parameter table: AIR Delay, AIR Diff Delay, AIR Multitap Delay, AIR Non-Lin Reverb, AIR Reverb, and AIR Spring Reverb. The other 17 are older Delay and Reverb algorithms (Delay Mono, Delay Ping Pong, Reverb Small, and so on) that share one line each with no individual writeup, covered in the "Not covered here" section below rather than guessed at.

Where these live: the biggest category

Every Delay or Reverb slot loads from the same shared list this site has now mapped across all six categories, one insert-or-send/return catalog, same four-slot structure everywhere it's used:

"Options: AIR Delay, AIR Diff Delay, AIR Multitap Delay, AIR Non-Lin Reverb, AIR Reverb, AIR Spring Reverb, Delay Analog Sync, Delay Analog, Delay HP, Delay LP, Delay Mono Sync, Delay Mono, Delay Multi-Tap, Delay Ping Pong, Delay Stereo, Sync (Stereo), Delay Tape Sync, Reverb In Gate, Reverb Large 2, Reverb Large, Reverb Medium, Reverb Out Gate, Reverb Small, Sample Delay"

Unlike EQ/Filter, where three of sixteen entries turned out to be genuinely comparable parametric equalizers, the Delay/Reverb category splits cleanly along a different line: discrete, countable echoes on one side, one continuous decaying tail on the other. AIR Delay, AIR Diff Delay, and AIR Multitap Delay are all delays, you can point at an individual repeat. AIR Non-Lin Reverb, AIR Reverb, and AIR Spring Reverb are all reverbs, one wash that fades rather than a series of taps.

AIR Delay: the clean one

Akai's own description leads with the one feature this effect has that the rest of the section largely doesn't, real stereo width control on the repeats themselves:

"This is a classic delay line effect with a variable feedback filter. Additional Ratio and Width parameters enable you to achieve a wide range of stereo delay effects."

This is the one page in the whole category that prints one parameter per line with its range and default sitting right next to it, no guessing required, and it's the page this article checked against Manual Version 3.7 field by field, identical on every row:

Parameter	Range	Default
Time (Sync Off)	1 ms – 2.00 s	388 ms
Time (Sync On)	1/32 – 8/4	1/8D
Sync	Off, On	Off
Feedback	0–100%	40%
Mix	0–100% (dry–wet)	50%
Delay Ratio	50:100 – 100:50	100:100
Delay HPF	20.0 Hz – 1.0 kHz	20.0 Hz
Delay Width	0–100%	100%
Feedback Damp	1.0 – 20.0 kHz	20.0 kHz
Feedback Reso	0–100%	0%
Fdbk. Reso Freq	100 Hz – 10.0 kHz	1.0 kHz

Feedback Reso and Fdbk. Reso Freq are a real pair, not two separate controls, Feedback Reso dials in how much resonance the feedback path gets and Fdbk. Reso Freq sets where that resonance sits, 0% Feedback Reso by default means the Freq field does nothing until you raise it. And Delay Ratio is the one field on this page that isn't obvious from its name, it splits the left and right repeat times against each other (50:100 through 100:50), not a stereo width control the way Delay Width is a separate, dedicated field just below it.

AIR Diff Delay: tempo-synced echoes

Diffusion is the word doing the work in this effect's own description, and it's not describing a room, it's describing how the repeats themselves smear:

"This is a delay line effect that is synchronized to your session tempo and uses an adjustable amount of diffusion to emulate the dissipation of echoes in reverberant space."

This page's Parameter, Range, and Default columns originally looked broken apart under a linear text pull, no reliable line-to-line pairing for the Default column, so this table first shipped ranges only. A position-sorted re-extraction of the same page (sorting each word by its own x/y coordinate instead of trusting the PDF's text-stream order) shows the three columns actually sit in clean, unbroken rows the whole way down, and every default below is cross-checked against v3.7 p.368, field for field identical:

Parameter	Range	Default
Time (Sync Off)	1 – 1000 ms	161 ms
Time (Sync On)	1/64 – 4/4, including triplet and dotted variations	1/16D
Sync	Off, On	On
Width	0–100%	100%
Mix	0–100% (dry–wet)	40%
Feedback	0–100%	50%
Fdbk. Diffusion	0–100%	40%
Fdbk. High Damp	0–100%	35%
Low Cut	20.0 Hz – 1.00 kHz	20.0 Hz
Pan	-100 – 0 – +100%	0%

Fdbk. Diffusion is the field that separates this effect from AIR Delay, it's own control over how much the repeats themselves scatter in the feedback path, on top of the tempo-synced Time field every delay in this category shares in some form. Defaults land toward the restrained end out of the box, 40% Mix, 50% Feedback, 40% Fdbk. Diffusion, not the washed-out setting the effect's own description might suggest.

AIR Multitap Delay: five independent taps

Where AIR Delay and AIR Diff Delay each produce one repeating echo, this effect is built to produce several at once, each with its own position:

"This effect is a versatile, creative delay plugin with a wide range of applications from subtle stereo delays to complex repeating patterns."

Same page, same fix: the position-sorted re-extraction resolves this table too, including a From/To field the original ranges-only table dropped entirely:

Parameter	Range	Default
Delay (Sync On)	1/16 – 8/4	4/4
Delay (Sync Off)	0 ms – 4.00 s	2.25 s
Feedback	0–100%	50%
Mix	-100.0 – 0.0 dB	-50.0 dB
From/To	Varies	Tap 5 / Input
Sync	Off, On	On
Low Cut	20.0 Hz – 1.00 kHz	500 Hz
High Cut	1.00 kHz – 20.0 kHz	2.00 kHz
Tap 1–5 (each)	Off, On	On
Tap Delay (each)	10.0 ms – 10.0 s	Varies
Pan (each)	L100 – C – R100	Varies
Level (each)	-Inf – 0.0 dB	0.0 dB

Five independent taps, each with its own On/Off, Tap Delay time, Pan position, and Level, is the real difference from the two single-repeat delays above. That's a genuinely different tool, one hit turning into a programmed pattern rather than a straight decaying echo. The From/To field the original table missed entirely picks which tap feeds which input, defaulting to Tap 5 pulling from the effect's own Input, worth knowing before assuming all five taps route the same way. Tap Delay, Pan, and Level all default to "Varies" per tap in Akai's own table, not a single shared number, the five taps ship pre-programmed with different values out of the box.

AIR Non-Lin Reverb: gated, not decaying

Every other reverb in this category is built to fade out naturally. This one is built to do the opposite, stop abruptly or play backward:

"This is a spatial effect, designed to produce synthetic, processed ambience with special gated and reversed reverb effects."

This page extracts cleanly under the same position-sorted read that resolves AIR Reverb and AIR Spring Reverb below: every row carries its own Parameter, Range, and Default on one line once sorted by page position, including the field the first, linear text-stream pass couldn't attribute safely. That first pass read one ambiguous "Pre-Delay" field with two stacked sub-values and left both numbers out rather than guess which was which. The position-sorted read shows they were never one field: Pre-Delay and Dry Delay are two separate, adjacent parameters, each with its own clean Range and Default. Shape also turns out to carry a listed default, Gated, missed entirely on the first pass:

Parameter	Range	Default
Pre-Delay	0–250 ms	0 ms
Dry Delay	0–1500 ms	0 ms
Time	0–1000 ms	250 ms
Mix	0–100% (dry–wet)	50%
Diffusion	0–100%	100%
Width	0–100%	50%
Shape	Gated, Reverse Gated	Gated
Low-Cut	20.0 Hz – 1.00 kHz	141 Hz
High-Cut	1.00 – 20.0 kHz	9.46 kHz

Shape is the field that actually delivers the "gated and reversed" behavior named in the description, a straight two-way switch between Gated and Reverse Gated, defaulting to Gated. Everything else on this page (Diffusion, Width, the two Cut filters) shapes the wash itself, Shape decides how that wash starts and stops. Pre-Delay and Dry Delay both gate the same silence-before-onset window, Pre-Delay controls the delay before the wet signal starts, Dry Delay is a separate, longer-range delay on the dry path, and both default to 0 ms out of the box.

AIR Reverb: 16 rooms, three tabs

This is the biggest single effect in the category by a wide margin, three sub-tabs (Reverb, Reverb Room, Hi/Lo Freq) where every other effect in this article gets one flat parameter list:

"This is a spatial effect, with a wide range of reverb types to add space or room to your audio signal."

The Type field's own option list extracted cleanly as a contiguous block even on the first pass, and it's worth quoting in full, sixteen named room and hall algorithms in one field:

"Off, Booth, Club, Room, Small Chamber, Medium Chamber, Large Chamber, Small Studio, Large Studio, Scoring Stage, Philharmonic, Concert Hall, Church, Opera House, Vintage 1, Vintage 2"

This article originally stopped at that Type list and Pre-Delay, the only two fields a linear text pull could attribute safely on a page this dense. A second pass reading the same page's word objects by their actual x/y position, then sorting top to bottom and left to right, keeps every Parameter next to its own Range and Default instead of letting the PDF's column order scramble them, and resolves the whole page, tab by tab:

Reverb tab
Parameter	Range	Default
Pre-Delay	0–250 ms	0 ms
Room Size	0–100%	100%
Time	0.4 ms – +inf s	1.9 s
Mix	0–100%	50%

Reverb Room tab
Parameter	Range	Default
Early Reflection Type	Off, Booth, Club, Room, Small/Medium/Large Chamber, Small/Large Studio, Scoring Stage, Philharmonic, Concert Hall, Church, Opera House, Vintage 1/2	Off
Early Reflection Length	0–100%	100%
ER/Tail Mix	0–100%	50%
Input Width	0–100%	0%
Output Width	0–100%	0%
Delay	0–250 ms	0 ms
Ambience	0–100%	0%
Density	0–100%	100%

Hi/Lo Freq tab
Parameter	Range	Default
Hi Freq Time	-100–0–100%	0%
Hi Freq Freq	2.00–20.0 kHz	6.32 kHz
Hi Freq Cut	1.0–20.0 kHz	9.46 kHz
Lo Freq Time	-100–0–100%	0%
Lo Freq Freq	20.0 Hz – 2.00 kHz	200 Hz
Lo Freq Cut	1–1000 Hz	1 Hz

Two numbers here are worth a second look because they match values already published earlier in this article: Hi Freq Cut defaults to 9.46 kHz and Lo Freq Cut's neighboring field bottoms out near the same 141 Hz–class range AIR Non-Lin Reverb's own High-Cut and Low-Cut fields default to above. That's not a coincidence this article is claiming as fact, just an internal consistency check, shared filter defaults across AIR effects that reuse the same underlying DSP block are plausible and nothing here contradicts a number already verified elsewhere in this article.

AIR Spring Reverb: the smallest page

Last of the six, and the shortest description in the category:

"This is a spatial effect, designed to emulate the sound of a spring reverb tank."

Same fix as AIR Reverb above closes this page too, six parameters, all now attributed with confidence:

Parameter	Range	Default
Pre-Delay	0–250 ms	3 ms
Time	1.0–10.0 s	4.0 s
Mix	0–100% (dry–wet)	50%
Diffusion	0–100%	100%
Width	0–100%	0%
Low Cut	20.0 Hz – 1.0 kHz	141 Hz

Side by side

Effect	Shape	What sets it apart	Table published
AIR Delay	Discrete taps	Delay Ratio splits L/R repeat time; real stereo Width	Full, cross-checked v3.7
AIR Diff Delay	Discrete taps	Fdbk. Diffusion scatters the repeats themselves	Full, cross-checked v3.7
AIR Multitap Delay	Discrete taps	5 independently panned, leveled taps	Full, cross-checked v3.7
AIR Non-Lin Reverb	Continuous tail	Gated / Reverse Gated, stops instead of decaying	Full (9 of 9 fields)
AIR Reverb	Continuous tail	16 named room/hall algorithms, 3 sub-tabs	Full, 3 tabs
AIR Spring Reverb	Continuous tail	Emulates a physical spring tank	Full

Not covered here: the other seventeen

The Delay/Reverb category's 23 entries include 17 that don't get an AIR name, a full description, or a dedicated parameter page in the manual, they share one line in the "Options:" list and nothing else: Delay Analog Sync, Delay Analog, Delay HP, Delay LP, Delay Mono Sync, Delay Mono, Delay Multi-Tap, Delay Ping Pong, Delay Stereo, Sync (Stereo), Delay Tape Sync, Reverb In Gate, Reverb Large 2, Reverb Large, Reverb Medium, Reverb Out Gate, Reverb Small, and Sample Delay. A few of these do get their own short entries later in the same appendix chapter (Delay Analog and Delay HP both showed up with real Parameter/Range/Default rows during this article's own research), but reconciling all 17 against the six AIR effects covered above is a bigger job than this article, disclosed as scoped out rather than rushed.

Common mistakes

Mistake: assuming every delay/reverb effect on this page got the same depth of coverage. It doesn't. AIR Delay, AIR Diff Delay, and AIR Multitap Delay are all fully verified and cross-checked against a second manual version. AIR Reverb, AIR Spring Reverb, and AIR Non-Lin Reverb are all fully resolved via a position-sorted re-extraction but not separately cross-checked against v3.7. Check the Side by side table above before assuming a specific range or default holds for a field not covered here.

Mistake: treating AIR Multitap Delay's Mix field like AIR Delay's. AIR Delay's Mix is a 0-100% dry/wet dial. AIR Multitap Delay's Mix is a -100.0 to 0.0 dB attenuation value, a different unit entirely, 0.0 dB is fully wet on this one, not the halfway point a percentage scale would suggest.

Mistake: assuming AIR Multitap Delay's five taps default to a uniform, straight repeating pattern. Tap Delay, Pan, and Level all default to "Varies" per tap in Akai's own table, not a single shared number, the five taps ship pre-programmed with different positions and levels out of the box, not zeroed uniformly.

A note on manual versions, and on what wasn't published

This site's own insert-vs-send-return article already confirmed the entire Effects & Parameters appendix carries a flat +21 page offset between Manual Version 3.7 and 3.9, Delay/Reverb specifically running v3.7 p.367 to v3.9 p.388, with every category's full "Options:" line matching word for word between versions. This article's own direct check of AIR Delay's parameter table against both versions confirmed the same thing at the individual-parameter level, not just the category name list, every one of AIR Delay's eleven rows identical between v3.7 p.367 and v3.9 p.388, including the specific 388 ms default.

That confidence doesn't carry evenly across the category, though two of the six turned out better than first believed. AIR Delay and, mostly, AIR Non-Lin Reverb print their parameter tables as one clean row per line, Parameter, Range, and Default all adjacent under a linear text pull. AIR Diff Delay and AIR Multitap Delay were originally read the same way as AIR Reverb and AIR Spring Reverb, Default Value seeming to print as a separate stacked block with no reliable pairing, and shipped ranges only rather than risk a wrong number. A position-sorted re-extraction, reading each page's text as individual word objects with their own x/y coordinates (PyMuPDF's get_text("words")) and sorting by position on the page, top to bottom then left to right, instead of trusting the order the text stream happens to emit them in, shows that original read was itself wrong: both pages' Parameter, Range, and Default columns actually sit in clean, unbroken rows the whole way down, no stacking at all. Cross-checked field by field against MPC_Standalone_OS_User_Guide_v3.7.pdf p.368, every value identical between manual versions, including a From/To row the original ranges-only table dropped entirely. This also corrected AIR Diff Delay's own page citation: its full description and table sit entirely on p.389, not p.388-389 as first cited, p.388 is the category's own chapter-intro page with no AIR Diff Delay content. AIR Reverb and AIR Spring Reverb had a different, real stacking problem this same technique also resolved (three sub-tabs, 18 rows, and six rows respectively, cleanly on the first attempt, not separately cross-checked against v3.7), the same failure mode this article's MPC1000 vs MPC2500 counterpart flagged in the MPC2500's own Specifications appendix and fixed with a second, independently-typeset source (ManualsLib) instead. The Type field's full sixteen-option list and the six confirmed field names published in this article's first version came from the same PDF; only the numeric attribution needed the second pass for AIR Reverb and AIR Spring Reverb. This also caught two page-citation errors in the article's original SOURCES.md: AIR Non-Lin Reverb's heading actually starts p.390, not p.389 as first cited, and AIR Spring Reverb's heading starts p.391, not p.390, both corrected in place.

A third update pass applied the same position-sorted technique to AIR Non-Lin Reverb's own page, still carrying two disclosed gaps at that point: a "Pre-Delay" field this article had read as one field with two stacked, unattributable sub-values, and a Shape field with no listed default. The position-sorted words show the "two sub-values" were never one field: Pre-Delay (0-250ms, defaulting 0ms) and Dry Delay (0-1500ms, defaulting 0ms) sit on their own separate lines with their own labels, a genuine two-field misread by the original linear pull, not an ambiguous PDF. Shape's default, Gated, was sitting on the same row as its Range the whole time, missed on the first pass along with the rest of that row's alignment. Both corrections came from the same PDF page already cited, no second source needed, and neither changes any number already published for AIR Reverb, AIR Spring Reverb, AIR Diff Delay, or AIR Multitap Delay. All six AIR effects in this category now carry a full parameter table.

Verdict

Spotted a pairing this article got wrong, on AIR Reverb, AIR Spring Reverb, or anywhere else in the category? Tell us about it.