审阅范围 · Review scope
限定范围独立审阅。请保留正文中的出处与限制；不代表全库最终验收、实际演奏或学习效果验证。
Independent review within the stated scope. Keep the evidence and limitations stated in the text in view. This is not final corpus acceptance or validation of performance or learning outcomes.
本页保留作者送审冻结字节，因此正文或数据可能仍有“候选／等待审阅／未发布”标记。当前版本已通过限定独立复核并发布；旧标记只记录冻结时状态。通过范围是双语含义、限定来源文字／静帧及原创固定例题检查，不含原声听校、连续动作、完整源谱、实际演奏、身体适用性或学习效果验证。
This page preserves the author’s frozen review bytes, so prose or data may still contain candidate, pending-review or unpublished labels. This version has passed bounded independent review and repository publication; those labels record author-time status only. Acceptance covers bilingual meaning, bounded source text/stills and original fixed-fixture checks, not source audition, continuous motion, full source scores, actual performance, physical suitability or learning outcomes.
同一个和弦，可以放在不同位置：先核音区、距离与三音
原创双语候选，待独立审阅，尚未发布。T2是有指定文字／静帧支持的作者观点；G是本例限定乐理；O是本库原创纸面任务。下列G和弦、时间表与答案不是源课谱例，也不是已验证音响效果的配置。
看到伴奏音隔得很远，先别急着张大手。把问题拆成三件事：用了哪些音名？它们具体有多高？先后顺序和时值怎样？相同和弦音换位置，可能仍保留原身份；三个音也可能只用了两种音名。这里先用纸面分清。
已读NC-053／054全部222段既有ASR并检查18次选定静帧，18个不同图像哈希。只取22段作两项简短作者归属的文字支持；其余200段不作归属证据，包括未使用／重复叙述、推广、不确定唱名与未验证的历史、音响或效果概括。这不等于200处已听校错误。选定谱面另有静态支持，完整源谱与原声未验收。
本页完整任务是纸面观察、找键编号和数拍，不要求钢琴、唱音、软件播放或宽跨度抓键。事件表没有指定手、指法或踏板，也没有可选弹奏路线。要把结果用于身体练习，须另作适用性、动作与音响检查；不能把这里的距离直接当作手掌跨度处方。
有边界的来源归属
B49-S01 · T2
第53课用移位排列说明三音仍可保留，并建议通常保留三音，同时明确容许有意省略。我们保留“先确认是否省音”的检查，不接受某种排列必然较好听，也不把作者偏好变成所有乐曲的规则。
既有ASR第15、17段约50.08–56.52秒，第40、42–43、46、49–54段约145.92–181.88秒；60／64／153秒选定谱音，174／179／191秒比较字幕。省略条件来自文字，静帧不能替代听校。
B49-S02 · T2
第54课建议比较同一和弦在不同起始音区和排列中的个人感受，并记录观察。我们把它转成先记清改变了什么的纸面准备；源课的轻巧、厚重等词是作者反应，不是用音程数值就能证明的声学结论。
既有ASR第99–108段约294.12–314.52秒；85／107／182秒选定八度位置，137／215／244／268／286秒选定间距，306秒作者描述表。137秒只显示两个已圈选音，未算作完整五音版本；没有听验这些排列。
先看身份，再看排列
G：音区是这一组音所在的高低范围；间距是相邻两个具体音高的距离。琶音／arpeggio把和弦音先后展开。这里“三音”从根音字母算第一位，取按字母顺序的第三位，不能与音列里的第三个项目混为一谈。用几次同名音，与用了几种和弦音，是不同计数。
G／O：以本页原创G大三和弦为基准，根音G、三音B、五音D。G–B是大三度，G–B♭是小三度。把B升高八度仍是B；删掉所有B才会使本页这条音列没有三音。只见G与D，不能在G大三和弦与G小三和弦之间二选一；更不能由此判定调性。甚至一个完整G大三和弦也不能独自判定整首乐曲的调。
G：本页采用科学音高记号，中央C写C4，紧邻下方的B是B3。标准88键钢琴从最低A0编号1，包含黑键，每移动到相邻键就是一个半音；C4为40，G3为35。编号只定位键，不规定手或指法。同一个字母的上八度相隔12键／半音。
G／O：可用这个纸面尺推算：自然音C–D–E–F–G–A–B–C的相邻距离依次为2、2、1、2、2、2、1半音；字母越过B到C才换八度编号。黑键也计入距离。从C4向下到B3只减1；整组降八度每个键号减12。先熟悉这个尺，再做下表；不会计算时保留为待核，不靠猜音名。
G／O：C→L整组降低一八度，保留内部距离；C→W只把B3移到B4，再按低到高列出，因此顺序和两个相邻距离都会变。W→N用G4替换B4，才使此音列失去B。不能把W的任何听感差异单独归因于“第一步变宽”：其他变量也变了，而且这里没有试听。
O：真实乐谱先保留原版各声部。伴奏省了B，不代表旋律、另一只手或持续声部也没有B；反过来，也不可凭和弦代号假定它已经响了。省略可能是有意编配，音区变化也不自动表示换和弦。这里的N只标明“G／D音名子集”，不替整段音乐决定和声功能。
原创纸面比较：哪一个真的少了三音？
先遮住下表编号、间距和身份三列，只看音列。用C4＝40、相邻键相差1推算编号；先写出每组用了哪些不计八度的音名，再算相邻键号之差。预测哪组没有B，哪两组的内部距离完全相同，哪一次改动同时改变了顺序与距离。最后揭开答案。
四组音列都按低到高排列；C／L／W／N只是版本ID，不是和弦代号。箭头表示读取顺序，不是同时抓键。间距写半音数，跨度是最后音减第一音；它们不是声学清晰度或手的可达性评分。
| ID | 音列 | 键号 | 间距／跨度（半音） | 身份 |
|---|---|---|---|---|
| C | G3 → B3 → D4 | 35, 39, 42 | 4, 3 / 7 | G大三和弦的全部三种音名 |
| L | G2 → B2 → D3 | 23, 27, 30 | 4, 3 / 7 | 同一音名集合，整组降八度 |
| W | G3 → D4 → B4 | 35, 42, 51 | 7, 9 / 16 | 同一音名集合，三音移高后重排 |
| N | G3 → D4 → G4 | 35, 42, 47 | 7, 5 / 12 | G／D子集，没有B；不是独立调性结论 |
核对答案
C与L的间距均为4、3，跨度7；它们相差整整一八度。W仍有G、D、B，三音在第三个项目，编号51；它的间距为7、9，跨度16。N有三个音，却只有G、D两种音名；第三个项目G4是根音的八度重复。N的间距7、5与跨度12都不能补回缺少的B。
若把N的第三项当三音，先分别圈“音列位置”和“和弦角色”；若把W误判为省略，擦去八度数字再比较音名集合；若L的距离改变了，逐个检查是否都减12。一次只修一个判断，再遮答案重做；没有因此证明能演奏或听辨。
保持时间不动，单独检查音高改动
下面是原创纸面时间表，不是播放器或演奏指令。四版均为3/4，四分音符＝60 BPM，两小节。表内从0开始计拍：三个音分别在0、1、2起音，在1、2、3结束；3–6是完整休止小节。先预测每版秒数、起音数，以及W→N是否需要改变时值，再看核对。
每格一拍，方括号是一小节；E1／E2／E3分别是三个事件ID，.为该条声部休止。不同版本复用ID表示对应位置，不把四版同时叠加。所有音是依次发生的单音，未分配左／右手；符号时值不保证某种实际连奏、断奏或余响。
[E1 E2 E3] [. . .]| ID | E1：0–1 | E2：1–2 | E3：2–3 | 休止：3–6 |
|---|---|---|---|---|
| C | G3 | B3 | D4 | 第二小节全部 |
| L | G2 | B2 | D3 | 第二小节全部 |
| W | G3 | D4 | B4 | 第二小节全部 |
| N | G3 | D4 | G4 | 第二小节全部 |
核对时间和恢复
每版共6拍＝6秒，第一小节3秒、第二小节3秒；3个起音、3个一拍音，后3拍休止。W→N只把E3的B4／51换成G4／47，时点不动。恢复W须把这个G4替换回B4，不能保留两音并称原样恢复；这会变成额外同时音，改变原事件。
稍后遮住答案，重建C、L、W、N并解释每个改动，作为保留重测。回到真实乐谱时，先核原音区、全部声部、音值与前后连接，再决定是否另列实验版本。可以记录“符号核对通过、听感／动作待核”。这些已公开的例子不是未见过的迁移测验；本包没有提供或验证迁移成效。
角色标签
root：根音
third：三音
fifth：五音
边界与相关阅读
来源账本合计636.075秒；同哈希MP4容器合计636.031秒，差0.044秒的原因未明，均不是听看时长。已核的是全部可用转写文字、选定静态记号以及原创符号计算，没有原声试听、连续动作、完整源谱、身体适用性、实际演奏、学习效果或应用验证。
本包来源范围
键号、集合与时间检查
One chord in different places: check register, spacing and the third
Original bilingual candidate, awaiting independent review and unpublished. T2 marks author views supported by specified text/stills; G theory scoped to this example; O our original paper tasks. The G-chord examples, schedules and answers below are not source scores or voicings with validated sonic effects.
When accompaniment notes look far apart, first separate three questions: which note names are used, how high are they, and in what order and durations do they occur? Moving chord tones can preserve their identity; three notes can also contain only two distinct names. We separate these on paper before considering movement.
All 222 existing ASR segments of NC-053/054 were read and 18 selected stills inspected, with 18 distinct image hashes. Only 22 segments support two brief attributions; the other 200 are unused as claim evidence, including unused/redundant narrative, promotion, uncertain solfege and unvalidated historical, sonic or efficacy generalizations. This does not mean 200 audition-confirmed errors. Selected notation has separate static support; complete source scores and original audio remain unvalidated.
The complete task here is paper observation, key-number lookup and beat counting. It requires no piano, singing, software playback or wide grip. The event table assigns no hand, fingering or pedal and offers no optional playing route. Physical application needs separate suitability, movement and sound checks; these distances are not hand-span prescriptions.
Bounded source attribution
B49-S01 · T2
Episode 53 illustrates retaining the third through relocation and generally recommends keeping it while explicitly allowing intentional omission. We retain the check for whether a tone was omitted, not a guaranteed sonic ranking or a rule for every piece.
Existing ASR segments 15/17 at about 50.08–56.52 seconds and 40/42–43/46/49–54 within 145.92–181.88; selected notation at 60/64/153 and comparison captions at 174/179/191 seconds. The omission condition comes from text; stills do not replace audition.
B49-S02 · T2
Episode 54 suggests comparing personal responses to one chord in different starting registers and arrangements, and recording observations. We adapt this into paper preparation that identifies what changed. Its light/heavy descriptions remain author responses, not acoustic conclusions proved by interval numbers.
Existing ASR segments 99–108 at about 294.12–314.52 seconds; selected octave positions at 85/107/182, spacings at 137/215/244/268/286 and author descriptions at 306 seconds. At 137 only two tones are circled, not an accepted complete five-tone version. None of the arrangements was auditioned.
Identity before arrangement
G: register is the high/low region occupied by the notes; spacing is the distance between successive specified pitches. An arpeggio unfolds chord tones successively. The chordal third uses the third letter position when the root is counted as first, not the third item in a sequence. The number of occurrences and the number of distinct chord tones are different counts.
G/O: our original G-major-triad baseline has root G, third B and fifth D. G–B is a major third; G–B♭ a minor third. Raising B by an octave retains B; removing every B leaves this sequence without its third. G and D alone cannot distinguish a G major from a G minor triad, still less establish a key. Even a complete G major triad does not by itself establish a whole piece’s key.
G: we use scientific pitch notation: middle C is C4 and B immediately below is B3. A standard 88-key piano numbers the lowest A0 as 1, includes black keys and advances one semitone per adjacent key. C4 is key 40 and G3 key 35. Numbers locate keys, not hands or fingers. The next octave of the same letter is 12 keys/semitones higher.
G/O: use this paper ruler: successive natural notes C–D–E–F–G–A–B–C are separated by 2,2,1,2,2,2,1 semitones. The octave number changes only across B→C. Black keys count in the distance. C4 down to B3 subtracts just 1; lowering a group by an octave subtracts 12 from every key. Learn this ruler before the table; mark an uncertain result pending rather than guessing names.
G/O: C→L lowers the entire group by an octave while preserving internal distances. C→W relocates only B3 to B4, then sorts low to high, changing order and both adjacent gaps. W→N replaces B4 with G4, removing B from this sequence. A sonic difference in W could not be attributed solely to a wider first step: other variables also changed, and no audition occurred here.
O: preserve every voice of a real score first. Omitting B from accompaniment does not remove B from melody, the other hand or a sustaining voice; conversely, a chord symbol does not prove B has sounded. Omission may be intentional, and register change does not automatically change the chord. N is only labeled a G/D pitch-name subset, not a verdict on the passage’s harmonic function.
Original paper comparison: which one actually lacks the third?
Hide the Key numbers, Gaps and Identity columns and inspect only the pitches. Use C4=40 and one unit per adjacent key to calculate numbers. List the note names without octaves, then subtract successive key numbers. Predict which group lacks B, which two preserve identical internal gaps, and which edit changes both order and spacing. Then reveal the answers.
All four sequences ascend. C/L/W/N are version IDs, not chord symbols. Arrows show reading order, not a simultaneous grip. Gaps are semitone counts; span is last pitch minus first. Neither measures acoustic clarity or hand reach.
| ID | Pitches | Key numbers | Gaps / span (semitones) | Identity |
|---|---|---|---|---|
| C | G3 → B3 → D4 | 35, 39, 42 | 4, 3 / 7 | All three pitch names of the G major triad |
| L | G2 → B2 → D3 | 23, 27, 30 | 4, 3 / 7 | Same name set, an octave lower |
| W | G3 → D4 → B4 | 35, 42, 51 | 7, 9 / 16 | Same name set, third relocated and reordered |
| N | G3 → D4 → G4 | 35, 42, 47 | 7, 5 / 12 | G/D subset without B; no independent key conclusion |
Check your answers
C and L each have gaps 4,3 and span 7, separated by exactly an octave. W still has G,D,B: its third is the third item, key 51; its gaps are 7,9 and span 16. N has three notes but only G,D as distinct names; its third item G4 duplicates the root an octave higher. Gaps 7,5 and span 12 cannot restore the missing B.
If you call N’s third item the chordal third, mark sequence position and chord role separately. If W looks like an omission, remove octave numbers and compare the name sets. If L’s gaps changed, check that every key number was reduced by 12. Correct one judgment, then retry with answers hidden; this does not demonstrate playing or aural recognition.
Keep time fixed while checking pitch edits
This is an original paper timeline, not a player or performance instruction. Every version has two bars of 3/4 at quarter note=60 BPM. Beat offsets start at 0: the three notes begin at 0,1,2 and end at 1,2,3; 3–6 is a complete rest bar. Predict duration, attack count and whether W→N changes timing before checking.
Each cell is one beat; brackets mark bars. E1/E2/E3 identify events and . a rest in this line. Reusing IDs across versions marks corresponding positions, not four simultaneous layers. Each note occurs successively, with no hand assignment; the written durations do not guarantee any particular articulation (legato or staccato) or pattern of acoustic decay.
[E1 E2 E3] [. . .]| ID | E1: 0–1 | E2: 1–2 | E3: 2–3 | Rest: 3–6 |
|---|---|---|---|---|
| C | G3 | B3 | D4 | whole second bar |
| L | G2 | B2 | D3 | whole second bar |
| W | G3 | D4 | B4 | whole second bar |
| N | G3 | D4 | G4 | whole second bar |
Check time and restoration
Each version lasts 6 beats=6 seconds: 3 seconds per bar, with three attacks, three one-beat notes and then three beats of rest. W→N changes only E3 from B4/key51 to G4/key47, not its times. Restoring W replaces that G4 with B4; retaining both adds a simultaneous tone and does not restore the original event.
Later, hide the answers, reconstruct C/L/W/N and explain each edit for a retention retest. On returning to a real score, check original registers, all voices, durations and joins before listing any experimental variant. Record “symbolic check passed; sound/movement pending.” These disclosed examples are not unseen transfer tests, and this package supplies or validates no transfer outcome.
Role labels
root：Root
third：Chordal third
fifth：Fifth
Limits and related reading
The source inventory totals 636.075 seconds and same-hash MP4 containers 636.031; the 0.044-second discrepancy is unexplained, and neither is viewing/audition time. Scope is all available transcript text, selected static symbols and original symbolic arithmetic—not original-audio audition, continuous movement, complete source scores, physical suitability, actual performance, learning efficacy or app validation.
Harmony and voicing boundaries
Separate retention from transfer
Source scope for this package
Key, set and timeline checks