Despite the current prevalence of phonological theory of dyslexia, there are several theories (e.g. the magnocellular hypothesis) that attribute an important role to visual deficits as a basis for dyslexia. These theories stem from introspective reports of many dyslexics of visual discomfort while reading and are further supported by findings of various visual deficits in dyslexic subjects. However, these findings were argued against and largely explained as resulting from impaired perceptual memory rather than poor immediate perception. To assess the role of (possibly impaired) visual perception in dyslexics' reading, we composed a task that was as similar as possible to normal reading in all its visual characteristics, but lacked all the other aspects of reading (phonological, semantic etc.), and compared performance of dyslexics and controls on it under several paradigms. The task was to identify a letter of an alphabet unknown to subjects, but similar to Hebrew and English in all graphical details (10 similar Georgian letters). Eight different conditions were assessed, measuring threshold duration of presentation (SOA) and threshold contrast levels for identification of small and large letters, with and without flanker letters, with and without white noise. Twenty adult native-Hebrew speaking dyslexics, mainly students, and 20 controls, matched for gender, age, and general cognitive abilities, participated in this study. We found all the predicted effects in both groups. Namely, adding flankers and decreasing letter size increased threshold SOAs, and adding white noise increased contrast thresholds. However, there was no difference between the experiment group and the controls, neither in single-set comparison, nor in effect magnitude, nor in an all-inclusive analysis of variance (MANOVA d=0; p>0.9). We conclude that the visual processing deficits found among dyslexic individuals by other researchers do not affect reading performance, and that therefore, the cause of their reading deficit resides elsewhere.