A Model for Automatic Optical Scaling of Type Designs for Conventional and Digital Technology
Total Page:16
File Type:pdf, Size:1020Kb
Rochester Institute of Technology RIT Scholar Works Theses 1987 A model for automatic optical scaling of type designs for conventional and digital technology Bridget Lynn Johnson Follow this and additional works at: https://scholarworks.rit.edu/theses Recommended Citation Johnson, Bridget Lynn, "A model for automatic optical scaling of type designs for conventional and digital technology" (1987). Thesis. Rochester Institute of Technology. Accessed from This Thesis is brought to you for free and open access by RIT Scholar Works. It has been accepted for inclusion in Theses by an authorized administrator of RIT Scholar Works. For more information, please contact [email protected]. A Model for Automatic Optical Scaling ofType Designs for Conventional and Digital Technology by Bridget Lynn Johnson A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in the School of Printing Management and Science in the College ofGraphic Arts and Photography of the Rochester Institute ofTechnology May 1987 Thesis Advisor: Professor A. Provan Certificateof Approval>- Master'sThesis Schoolof Printing Rochester Institute of Technology Rochester, New York CERTIFICATE OF APPROVAL MASTER'S THESIS This is to certify that the Master's Thesisof MASTER'S THESIS This is to certify that the Master's Thesisof Bridget Lynn Johnson nameof student With a major in Printing Technology has been approvedby the ThesisCommittee as satisfactory for the thesis requirement for the MasterofScience at the convocation of MrlV lqR7 1 ABSTRACf In the history of type design, two methods have been used to scale type-to produce enlarged or reduced letterforms from a reference size. With original handcut fonts, designers performed opticalscaling (scaling by eye) that varied the proportions of letterform features over a range ofsizes in a nonlinear manner. That is, letterform feature proponions were size dependent. This was an entirely manual and intuitive process. More recently, however, the use of the lens, as well as computational and other technologies, has allowed letterforms to be scaled automatically from a reference character, a simple proportional enlargement or reduction. To date, little work has been done to combine these two methods, that is to say, to automatically perform nonlinear scaling of a reference character in order to approximate the optical scaling performed by skilled type designers and punchcutters. This research developed a mathematical model ofoptical scale in type design, consisting of two parts: (1) a model ofthe scaling ofindividualletterform features; and (2) a model of the scaling ofentire letterforms. The model was tested by applying it to the original handcut fonts that supplied the initial data for the research in order to generate synthetic letterforms. These nonlinear synthetic letterforms were then compared with the originals, as well with proponionally scaled letterforms generated from the originals. The goal was to determine how well the nonlinear letterforms generated by the model approximated the original optically scaled handcut letterforms. In addition, the performance of the proportionally scaled letterforms was compared with the originals, as well as with the nonlinear, synthetic forms. TABLE OF CONTENTS LIST OF FIGURES . iv CHAPTER I. INTRODUCTION 1 Hand Punchcutting (2)--Mechanical Punchcutting (6)--Phototypesetting (9)- Digital Typography (10) CHAPTER II. LITERATURE REVIEW 14 CHAPTER III. FONT HISTORY 19 Caslon (20)--Clarendon (25)--Bodoni (28)--Akzidenz Grotesque (32) CHAPTER IV. HYPOTHESES 36 CHAPTERV. METHODOLOGY 39 Modeling the Scaling of Individual Letterfonn Features (39)--Modeling the Scaling of Entire Letterfonns (41)--Testing the Letterfonn Scaling Model (44) CHAPTER VI. RESULTS OF RESEARCH 46 CHAPTER VII. ANALYSIS OF DATA AND DISCUSSION OF RESULTS 71 Scaling of Individual Letterform Features (71)--Letterforms Resulting from the Application of the Nonlinear Scaling Model (76)--Font Review (78)- Application of Model to Other Fonts (87)--Low Resolution (92) CHAPTER VIII. CONCLUSIONS . 96 CHAPTER IX. RECOMMENDATIONS FOR FURTHER INVESTIGATION 99 11 BIBLIOGRAPHY 101 APPENDIX I. NUMERICAL ANALYSIS 104 APPENDIX II. FEATURE EQUATIONS 109 APPENDIX III. SCALING DATA . 115 APPENDIX IV. FEATURE GRAPHS 121 APPENDIX V. LETTERFORM NOMENCLATURE 134 111 LIST OF FIGURES 1. A hand punchcutter at his workbench: 1. handcut punch, 2. striking the matrix, 3. unjustified strike, 4. squared matrix ready for casting, 5. cast type . ., 4 2. Lanston Monotype pantographic mechanical punchcutting machine with a close-up ofa brass relief pattern and the cutting tool 8 3. Times Roman R and a in high and low resolutions 13 4. William Caslon and Sons 1734 Specimen Sheet 21 5. Exaggerated English Modem style . 22 6. Caslon, Dutch Old Face, English handwriting, and Baskerville original handcut specimens 24 7. Clarendons vs. Egyptians 27 8. Specimen of Bodoni . 29 9. Specimen of Akzidenz Grotesque . 35 10. Example of a feature plot with sample data 40 11. Visual display ofhorizontal growing and scaling . 43 12. (Caslon CWO) Comparison of synthetic letterforms with handcut letters 47 13. (Caslon EHR) Enlargements of synthetic and original sizes at high resolution 48 14. (Caslon ELR) Enlargements ofsynthetic and original sizes, low resolution 49 15. (Caslon RS) Full size range ofsynthetic letterforms at high and low resolutions. .. ., 50 16. (Clarendon CWO) Comparison ofsmall, medium and large sizes of synthetic letterforms with original handcut letters . 51 17. (Clarendon EHR) Enlargements of synthetic and original sizes, high resolution ... .. 52 18. (Clarendon ELR) Enlargements of synthetic and original sizes, low resolution . 53 19. (Clarendon RS) Full size range of synthetic letterforms at high and low res. 54 20. (Bodoni CWO) Comparison ofsmall, medium and large sizes of synthetic letterforms with original handcut letters ... 55 21. (Bodoni EHR) Enlargements of synthetic and original sizes, high resolution 56 22. (Bodoni ELR) Enlargements ofsynthetic and original sizes, low resolution 57 23. (Bodoni RS) Full size range of synthetic letterforms at high and low res. 58 24. (Akzidenz CWO) Comparison of small, medium and large sizes of synthetic letterforms with original handcut letters . 59 25. (Akzidenz EHR) Enlargements of synthetic and original sizes, high resolution . 60 26. (Akzidenz ELR) Enlargements of synthetic and original sizes, low resolution 61 27. (Akzidenz RS) Full size range of synthetic letterforms at high and low res. 62 28. (Times Roman CWO) Comparison ofsmall, medium and large sizes of synthetic letterforms with original handcut letters . 63 29.:(Times Roman EHR) Enlargements ofsynthetic and original sizes, high res. 64 tv 30. (Times Roman ELR) Enlargements of synthetic and original sizes, low res. 65 31. (Times Roman RS) Full size range of synthetic letterforms at high and low res.. .. ., 66 32. (Helvetica CWO) Comparison of small, medium and large sizes of syntheticletterformswith original handcutletters . 67 33. (Helvetica EHR) Enlargements of syntheticand original sizes, high resolution . 68 34. (Helvetica ELR) Enlargements of synthetic and original sizes, low resolution 69 35. (Helvetica RS) Full size range of synthetic letterformsat high and low res. 70 36. Feature equation plot showing nonlineareffect within the size range 6 to 36 points. ., 74 37. Range ofresolution for which the nonlinearscaling model can be applied using the values of Bv and Bh 93 38. Classificationsof horizontaland verticalmeasurements 105 39. Implementation of scale and growth factors 106 40. Upper case serif width 122 41. Upper case seriflength 123 42. Lower case serif length 124 43. Upper case stem width 125 44. Lower case stem width 126 45. X-height . 127 46. Ascender height . 128 47. Descender height. 129 48. Lower case 0 bowl width 130 49. Capital letter 0 width 131 50. Capital letter H width. ... 132 51. Horizontal and vertical scale and growthparameters 133 52. Capital Letterform Nomenclature . 135 53. Capital Letterform Nomenclature (continued) 136 54. Lower Case Letterform Nomenclature. 137 55. Lower Case Letterform Nomenclature (continued) 138 56. Ligature and Serif Nomenclature 139 57. Additional Nomenclature 140 v 1 CHAPTERI INTRODUCTION Handwritten letterformscome in a variety of sizes for a given design. This practice was carried over from the scribesinto the early productionof printing types. As printing and the manufacturing of printing typesevolved, type became availablein a wider range of sizes and designs. Printing types which were manufacturedby hand had certain design features that were size dependent For a particulartypeface in a given size, individual feature parts had subtly differentproportionsto compensatefor the size at which they were manufactured. These optical compensations produced an overall consistencyin design and legibilitythrough a range of sizes and later becameknown as optical scalein type design. In the last one hundredyears there have been advancesin the design and manufacturing of printing types towards automatically scaling letterforms. This involves converting a drawing of a givencharacter(the reference character) into anothersize with the same or similar proportions to the original.Iffeaturesremain a constant proportion of the original reference image, this practiceis known as proportional scaling. Another possibilityexists to do automatic optical scaling that approximates what hand punchcutters practiced when they did everything by eye and by hand. This would