History
From Chromium Bug tracker and dialogs
Adobe include special hints stream into their linerilized PDFs. Format of such stream is black box and understood only by Adobe software. This allows Adobe to load linerilized PDFs more efficiently.
The alternative is server partial byte ranges. Chromium Bugs for read and alter were opened 2011
reputedly in 2014 Adobe posted (source no longer on Adobe site) along the lines of indicate that Google chrome does not support byteserving a PDF. That bug is still open whilst others related were closed "Wont fix"
Thus both methods are problematic for Foxit/Skia (Chromes binary developers) and other 3rd Parties to emulate.
try this 2GB Link in Acrobat it should open page 1 before download. Chrome needs to wait for the whole download ! then although Downloaded! My Edge expects to save it in about 10 minutes time.
https://s3.amazonaws.com/pdftron/downloads/pl/2gb-sample-file.pdf
One commercial solution :- (A few seconds to display their file)
PDFTron (owner company of Iceni, iText, PDFLib, XODO and many others) PDF.js more specifically a "Pre-Loaded" Express can
Firefox/Mozilla using simpler JavaScript methodologies
It can vary if Firefox viewer can use Linearized files, one of my older legacy viewers does not work with that sample, but the newer inbuilt FireFox enabler does.
Unclear if either can be passed through Mozilla PDF.js based React PDF Viewer, but supposedly was, from 3.3.3 onwards. However the dev viewer as linked above, does not exhibit that behaviour when using simple
unserved file open e.g. "drag and drop" However that 1st Page does appear fast (along with others) after the time taken to upload.
NOTE a "/Linearized" PDF file must not have been "adapted" e.g. annotated or simply served (other than as transcoded text) etc. after "Web Optimisation". The viewer must be working with the decimal addressed binary file bytes from server.