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This page preserves the author’s frozen review bytes, so prose or data may still contain candidate, pending-review or unpublished labels.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.
同一和弦，换排列：把分解伴奏练成可核对的小任务
作者送审冻结版本；当前发布与审阅状态由仓库导航说明。
先把伴奏当成一个小项目
先保留同一组音和同一条时间线，只比较音的次序。这里的成功标准是能指出哪里改变、哪里保持，并能把缩短练习准确接回完整版本；“更好听”不作为纸面答案。
先看得懂，再弹
和弦是一组有关系的音；分解和弦把这些音先后安排。想像把同一套积木排成两列：积木可以相同，次序却不同。本页设定Dm（D小三和弦）为D–F–A，三音先后出现，不要求同时按住。这个设定不证明整首曲子的调或功能。
三个音与三个位置
本页用科学音高记法：中央C为C4；其右侧相邻白键是D4，再向右越过E4到F4，越过G4到A4。数字4是音区，不是指法。88键按最左A0为1数，D4/F4/A4为42/45/49；MIDI编号分别62/65/69，与88键编号不是同一套。
低、中、高不等于根、三、五
在本页指定的D4–F4–A4配置中，低／中／高恰好是根音D、三音F、五音A。若改成F4–A4–D5，仍是给定Dm的音，最低却是F。低高中高描述实际高低顺序，不能照搬成任何配置的根五三五，也不是左手指法。
阿尔贝蒂低音的一个结构入口
低—高—中—高是阿尔贝蒂低音（Alberti bass）常见的四音排列。本页B用它做孤立的入门音序；没有旋律、低音声部配器或历史作品语境，不把这个练习称为一首完整伴奏。重复最高音不自动要求加大触键力度或更响。
先读共同条件
A、B、R各自独立：4/4，四分音符=60 BPM，一拍1秒，一小节4秒，总长两小节8秒。偏移0表示开始，偏移4是第二小节开头。A/B的第一小节有8个八分音符，每音占0.5拍=0.5秒；第二小节整小节休止。没有速度、强弱、踏板或手分工的隐含变化。
怎样读事件表
一行表示一个新音开始和记号上的结束；[0,0.5)包含起点，不包含终点。前音结束与后音开始可同在0.5。两个相邻A4行仍是两次起音，不是延音线。表格规定的时间不能保证实际连奏、断奏、声音长度或琴弦余响。
先预测，再核对
先遮住答案，仅看下面的事件表：A和B是否用了同样的音与同样的重复次数？每个四音组中，哪些位置不同？它们是换和弦还是换排列？R少了哪些起音，哪些空档必须保留？先记录你的答案，再展开核对。此处是已给定材料的练习，不是陌生乐谱迁移测试。
偏移单位：四分音符拍；60 BPM时数值也等于秒。
| Version / 版本 | Event / 事件 | Pitch / 音 | Start / 起点 | End / 终点 |
|---|---|---|---|---|
| A | A-N1 | D4 | 0 | 0.5 |
| A | A-N2 | F4 | 0.5 | 1 |
| A | A-N3 | A4 | 1 | 1.5 |
| A | A-N4 | A4 | 1.5 | 2 |
| A | A-N5 | D4 | 2 | 2.5 |
| A | A-N6 | F4 | 2.5 | 3 |
| A | A-N7 | A4 | 3 | 3.5 |
| A | A-N8 | A4 | 3.5 | 4 |
| A | Rest / 休止 | — | 4 | 8 |
| B | B-N1 | D4 | 0 | 0.5 |
| B | B-N2 | A4 | 0.5 | 1 |
| B | B-N3 | F4 | 1 | 1.5 |
| B | B-N4 | A4 | 1.5 | 2 |
| B | B-N5 | D4 | 2 | 2.5 |
| B | B-N6 | A4 | 2.5 | 3 |
| B | B-N7 | F4 | 3 | 3.5 |
| B | B-N8 | A4 | 3.5 | 4 |
| B | Rest / 休止 | — | 4 | 8 |
| R | R-N1 | D4 | 0 | 0.5 |
| R | R-N2 | A4 | 0.5 | 1 |
| R | R-N3 | F4 | 1 | 1.5 |
| R | R-N4 | A4 | 1.5 | 2 |
| R | Rest / 休止 | — | 2 | 8 |
核对音序与时间
A每组为D4–F4–A4–A4，B为D4–A4–F4–A4，两者均重复两组。每组只交换第2、3个位置；第1、4个位置相同。每版D4两次、F4两次、A4四次，都有8个不同起音时点，所用音高集合相同。在给定Dm标签下，这次没有换和弦、音区或速度。
缩短练习必须留下空档
R只保留B的前四个事件，0–2秒照原样；从2秒到8秒休止。它仍是两小节，不把剩余音挤到前面，也不把第二小节提早。还原时补回偏移2、2.5、3、3.5的D4、A4、F4、A4，并把休止改回4–8。不要只补音而保留重叠的2–8休止。
一个小任务怎样验收
任务包：先在纸上圈出每组低／中／高，写出两个交换位置，再核每音起止和休止。完成证据是一张正确标记的A/B对照与一次精确的R→B还原；错误就定位到一个音位或一个时点，不一次改音序、速度和力度。重复次数是你记录的结果，不是保证学会的固定处方。
可选的单手核对
若能舒适地找到D4、F4、A4，可任选一只手按事件表试B，再试R→B；无需同时伸展按住三个音，不规定指法或手形，不踩延音踏板。按每行结束时放开该键，下一行再起；同键重复也要重新起音。若无法保持舒适或不能稳定辨认按键，就回到纸面或请教师帮助。
速度降低时写明新目标
若实际核对需更慢，可把四分音符改成30 BPM；音值、音序与两小节结构保持，每拍2秒、每音1秒、全版16秒。表中偏移仍以四分音符拍为单位；不能沿用60 BPM的秒数。先写清目标再试，不要求达到任何统一“达标速度”。
把反馈变成下一步
若第二个音仍是F4，先在B上圈出“高=A4”，只读一个四音组；若连续A4被当成长音，数两次起音；若R休止被跳过，边数完两小节边指时间线。实际弹奏时，可记录自己听见的次序或节拍偏差；这些观察不能确定手指动作或生理原因。一次只选一个可核问题。
隔开再测，再放回乐句
做另一项任务后，遮住答案再标A/B并还原R；记录间隔和结果，不用当场熟练代替保留。将来用于自己的合法乐谱时，先核原谱音序、和弦、实际低音、节拍、表情和旋律，再练一个短段及前后连接；忠实练谱时不能擅自把原型换成B。新的陌生题需另行设计，今天的重复不等于迁移证据。
教材与来源的边界
这是一份原创控制变量练习；A不是作者示例，B只采用一般低高中高结构，R和验收流程也是本库设计。所选源片段只用于有限文字／静帧归属。没有原声听校、连续动作、完整源谱、实际演奏、身体适用性或学习效果验收；完整来源课程仍不能标成已看完。
来源与继续阅读
NC-009在ASR段36–39及201秒静帧介绍低—高—中—高。NC-031在ASR段9、10、17、18及49／54秒静帧区分和弦配置与伴奏起音；这是作者的编配思路，不是必须遵守的练谱法。本页A/B/R与验收步骤是原创设计。段号从0计，秒数是选帧位置，不是已听时长。
Same chord, different order: a checkable broken-chord task
Author-time frozen submission; repository navigation states current publication and review status.
Treat accompaniment as a small project
Keep the pitch collection and timeline fixed, then compare note order. Success here means identifying what changed and what stayed, and restoring a shortened exercise to the complete version. “Sounds better” is not a paper answer.
Understand it before playing
A chord is a related group of pitches; a broken-chord pattern places them in sequence. Like arranging the same blocks in two rows, the materials can stay while their order changes. Here Dm (D minor triad) is stipulated as D–F–A, presented successively without requiring simultaneous holding. This does not establish a whole piece’s key or harmonic function.
Three notes and three positions
Scientific pitch notation is used: middle C is C4; the adjacent white key to its right is D4, then pass E4 to F4 and G4 to A4. The numeral 4 is register, not fingering. Numbering an 88-key piano from leftmost A0=1 gives D4/F4/A4 as 42/45/49. Their MIDI numbers are 62/65/69, a different numbering system.
Low, middle and high are not always root, third and fifth
In this specified D4–F4–A4 voicing, low/middle/high happen to be root D, third F and fifth A. With F4–A4–D5, the pitches still belong to the given Dm but F is lowest. Low–high–middle–high describes actual pitch order, not root–fifth–third–fifth in every voicing or a left-hand fingering.
A structural entry to Alberti bass
Low–high–middle–high is a common four-note ordering in Alberti bass. B uses it as an isolated introductory pitch sequence. Without a melody, bass-part arrangement or historical-work context, this exercise is not a complete accompaniment. Repeating the highest pitch does not automatically require more force or loudness.
Read the shared conditions first
A, B and R are separate versions: 4/4, quarter note=60 BPM, one second per beat, four seconds per bar, two bars totaling eight seconds. Offset 0 is the start and offset 4 begins bar 2. A/B have eight eighth notes in bar 1, each lasting 0.5 beat=0.5 second; bar 2 is a full-bar rest. There are no hidden changes of tempo, dynamics, pedal or hand allocation.
How to read the event table
Each row gives a new note’s onset and symbolic end; [0,0.5) includes the start but not the end. One note can end as the next begins at 0.5. Adjacent A4 rows still mean two attacks, not a tie. The timing grid cannot guarantee actual legato, staccato, sound duration or string resonance.
Predict before checking
Hide the answers and use the event table below: do A and B use the same pitches with the same repetition counts? Which positions differ in each four-note group? Is this a chord change or an ordering change? Which attacks does R omit, and which gaps must remain? Record your answers before checking. This is practice on disclosed materials, not an unseen-score transfer test.
Offsets are quarter-note beats; at 60 BPM their values also equal seconds.
| Version / 版本 | Event / 事件 | Pitch / 音 | Start / 起点 | End / 终点 |
|---|---|---|---|---|
| A | A-N1 | D4 | 0 | 0.5 |
| A | A-N2 | F4 | 0.5 | 1 |
| A | A-N3 | A4 | 1 | 1.5 |
| A | A-N4 | A4 | 1.5 | 2 |
| A | A-N5 | D4 | 2 | 2.5 |
| A | A-N6 | F4 | 2.5 | 3 |
| A | A-N7 | A4 | 3 | 3.5 |
| A | A-N8 | A4 | 3.5 | 4 |
| A | Rest / 休止 | — | 4 | 8 |
| B | B-N1 | D4 | 0 | 0.5 |
| B | B-N2 | A4 | 0.5 | 1 |
| B | B-N3 | F4 | 1 | 1.5 |
| B | B-N4 | A4 | 1.5 | 2 |
| B | B-N5 | D4 | 2 | 2.5 |
| B | B-N6 | A4 | 2.5 | 3 |
| B | B-N7 | F4 | 3 | 3.5 |
| B | B-N8 | A4 | 3.5 | 4 |
| B | Rest / 休止 | — | 4 | 8 |
| R | R-N1 | D4 | 0 | 0.5 |
| R | R-N2 | A4 | 0.5 | 1 |
| R | R-N3 | F4 | 1 | 1.5 |
| R | R-N4 | A4 | 1.5 | 2 |
| R | Rest / 休止 | — | 2 | 8 |
Check order and time
Each A group is D4–F4–A4–A4 and each B group is D4–A4–F4–A4, repeated twice. Only positions 2 and 3 swap in each group; positions 1 and 4 match. Each version has two D4s, two F4s, four A4s and eight distinct attack times, using the same pitch set. Under the given Dm label, this changes neither chord, register nor tempo.
A shortened exercise must preserve its gap
R retains exactly B’s first four events at 0–2 seconds, then rests from 2 to 8. It remains two bars: do not compress the remaining notes or bring bar 2 forward. Restore D4, A4, F4 and A4 at offsets 2, 2.5, 3 and 3.5 and change the rest back to 4–8. Adding notes while retaining an overlapping 2–8 rest is not restoration.
How to finish one small task
Work package: label low/middle/high on paper, identify the two swapped positions, then check every onset, end and rest. Completion evidence is a correctly marked A/B comparison and one exact R→B restoration. Locate an error to a pitch position or time point instead of changing order, tempo and loudness at once. Repetition counts are observations you record, not a fixed guarantee of learning.
Optional one-hand check
If you can comfortably locate D4, F4 and A4, choose either hand to try B and then R→B from the grid. No simultaneous three-note stretch, fingering or hand shape is prescribed; leave the sustain pedal unused. Release the key at each row’s end and start the next row; repeated keys require a new attack too. If comfort or reliable key identification is lacking, return to paper or seek a teacher’s help.
State the new target when slowing down
If the optional check needs a slower pace, use quarter note=30 BPM, preserving values, order and the two-bar form: two seconds per beat, one second per note and sixteen seconds per version. Table offsets remain in quarter-note beats; do not retain the 60-BPM seconds. Write the target before trying; there is no universal required passing tempo.
Turn feedback into a next action
If the second note remains F4, circle “high=A4” in B and read just one four-note group. If consecutive A4s become one long note, count two attacks. If R’s rest is skipped, point along the timeline while counting both bars. During optional playing, record order or pulse differences you notice; these observations cannot establish finger motion or a physiological cause. Choose one checkable issue at a time.
Retest later, then return to the phrase
After another activity, hide the answers, annotate A/B again and restore R; record the delay and result instead of treating immediate fluency as retention. For your own lawful score, first check its note order, harmony, actual bass, meter, expression and melody, then practice a short segment and its joins. Faithful score practice does not license replacing the written pattern with B. New unseen tasks need separate design; today’s repetition is not transfer evidence.
Lesson and source boundaries
This is an original controlled comparison. A is not an author example; B uses only the general low–high–middle–high structure; R and the completion procedure are also our design. Selected source passages support bounded text/still attribution. There has been no source audition, continuous-motion review, complete source-score audit, performance, physical-suitability or learning-outcome validation; the source courses cannot be marked fully watched.
Sources and next reading
NC-009 introduces low–high–middle–high in ASR segments 36–39 and the 201s frame. NC-031 distinguishes voicing from accompaniment attacks in segments 9, 10, 17 and 18 and frames at 49/54s. This is the author’s arranging strategy, not a mandatory score-practice method. A/B/R and the completion steps are original designs. Segment IDs are zero-based; seconds locate stills, not duration heard.
Source scope and concise attributions · Original event data · Fixed-fixture checks · Register and chord identity · Meter and attacks