Jump to content


  • Posts

  • Joined

  • Last visited

  • Donations


Everything posted by bob2000

  1. Thanks for the info. I found this thread http://www.msfn.org/board/index.php?showto...55706&st=40 but all the links to IconViewer have expired. Could you please re-upload it somewhere else ?
  2. Hi WildBill, What version of Xpize did you use to patch Windows 2000 ? I'm asking because support for Windows 2000 isn't mentioned anywhere (http://www.xpize.net/download.php), so I guess it must have been an older version (they are all still available, luckily). Also, can this be installed on a running machine (e.g. as a patch) or does it have to be necessarily slipstreamed onto the OS install CD ? I realize this isn't the most appropriate thread for these questions, but it'd be great if you could answer anyway. Thanks a lot.
  3. Daedalus is great ! The only bug I have found is that IconTweaker will not start (or run properly) if the Deadalus service is running. So when I want to use IconTweaker, I must stop the service first, make the icon changes, and then re-start the service (and sometimes the shell icon cache needs to be cleared). This happens on Windows 2000 Professional and Windows 2000 Server. Apologies if this problem has already been reported: seeing as IconTweaker is what many use to update their system icons, I though I'd mention this issue.
  4. Great stuff ! I can't wait to try out the new skinning features. Of all the functionality that might/will be re-used from TClock3, the thing I'm most looking forward to is the transparent background for desktop icon labels and the outline/shadow around the icon text. Speaking of icons, at some point it would be great to see Daedalus' 32-bit icons support integrated into SmoothText (as you suggested yourself in that thread, WildBill). That would make SmoothText a great, all-in-one styling software
  5. I have tried SmoothText 0.7.9 on Windows XP Professional, SP2. I found 3 problems with the taskbar translucency: - there are no minimized windows on the taskbar, i.e. the taskbar is empty - clicking on the Start menu, the menu appears invisible, although it still casts a shadow, and it actually still works, e.g. if you hover where the "My Recent Documents" entry normally is, you do get the proper, visible My Recent Documents menu. - after a little while, this made explorer.exe crash (may or may not be an XP-specific issue, seeing Colonel O'Neill's last post)
  6. Hey WildBill, Remember my post about 10pt Arial being distorted vertically, with lower case letters being bigger than they shoud be? (post #147). You mentioned Windows' internal hinting as a possible explanation. Well, I found that if you substitute Arial with Arial Unicode MS in SmoothText, there's no distortion in the rendered text. One obvious explanation is that Arial and Arial Unicode MS aren't perfectly identical, as Arial Unicode MS tends to appear just slightly wider: http://en.wikipedia.org/wiki/Arial_Unicode_MS But for most font sizes, the two look very very similar, and it's impossible to tell them apart (except of course for the symbols that are altogether different). I'm not sure if this helps at all, but I thought I'd mention it anyway. Other than that, I'm really looking forward to SmoothText 0.7.9 and all the other skinning features that will follow
  7. Thanks a lot for the info. I'm aware of SmoothText, and I have been using it for some time now. Lately I have also been giving WildBill my small contribution, with a few suggestions on how to expand SmoothText. What caught my attention in this thread is Tihiy's claim that this is the real ClearType, which sounds very promising. That made me wonder if the same approach could be adopted on Win2000.
  8. Hi WildBill, Here's an idea of how SmoothText could be customised even further (hopefully without confusing any user ): I'm thinking of Tuning profiles. Let's assume that SmoothText: 1) knows which application it's currently drawing text for (e.g. Explorer, Firefox, Word, etc.) 2) can change Tuning parameters (Energy distribution, Gamma, Width percent/bias, Force character placement) on the fly, with an acceptable performance hit. If either of the above isn't true, then you can disregard most of this post... On the other hand, if both assumptions are true, Tuning profiles would work like this. The Tuning tab would have a pull-down menu at the very top, specifying the choice of profile. Only two entries would be there, initially: "Default" (the catch-all profile that by default applies to all applications, unless specified otherwise in another profile) and "New profile..." (to create a new profile). When a new profile is created, the Tuning tab would have an Include section, with two lists of applications side-by-side (the one on the left, full, the other, empty, initially) and "Add >>" / "<< Remove" buttons in between, in the same way that the Problem fonts tab already has at the moment. Bottom line: when drawing text for a given application, e.g. Firefox, if that application appears in a profile, the Tuning settings defined for that profile are used. Otherwise, the default Tuning settings are used. [note: 1) when creating a profile, checks would ensure that every chosen application doesn't appear in more than one profile 2) applications can only be associated with non-Default profiles] Standard users would continue to use the Default profile for everything, and you wouldn't be introducing any mandatory extra configuration. Advanced users would be able to customise the text appearance as much as they want to. I realise that, even assuming all of this CAN be done, it would take quite a lot of coding effort. But I think the benefits would be huge. I know I'd be creating straightaway one profile for Explorer, one covering all web browsers and one for word processors. This is because, in my opinion, Tuning settings which are ideal for one use very often are not good at all for others. E.g. With the best settings for web browsing (E10 / G1.40, Forced character placement On), text on Explorer looks very thin, white-ish and irregularly spaced. Instead, Explorer looks great with E6 / G1.00 and Forced character placement Off. And choosing the best compromise settings just doesn't do either application justice, in my opinion.
  9. Tihiy, all, Does this work on Windows 2000 by any chance? With the Microsoft Reader supporting all OSs, I would imagine it all depends on the new DLL.
  10. The new option to force characters into the original cells works very well: in my opinion, it removes most of the font-width issues and it improves the overall quality of the text, because of the more regular spacing. And I have yet to find any major, undesired side-effect, so for me this option is definitely staying on As for the witdh percent and width bias, I find that the default value is best for both. Which brings me to one last rendering issue. SmoothText distorts 10pt Arial in its vertical proportions, which also means all its lowercase letters are taller. These are screenshots of eBay's Categories menu (obviously, 10pt Arial is used a lot also elsewhere e.g. Google, Yahoo! Mail). The first one is the Windows 2000 default text. The second one is SmoothText 0.7.6, Energy 10, Gamma 1.30, Force character placement On, with default values for Width percent and Bias (compare the "r" and "t" in "Art" with the original text, for example). I guess this isn't expected behaviour, as the sub-pixel anti-aliasing only operates horizontally, and shouldn't impact vertical proportions. The only way to fix this is to set the Width bias to its rightmost, Wide setting. This, unfortunately, messes up the alignment of a lot of websites (e.g. a typical news site, spanning multiple columns and using several styles), so it isn't a viable setting for me. Note that with the Width bias set to any value other than its rightmost Wide setting, the vertical distorion re-appears.
  11. Hi WildBill, Thanks a lot for sharing the implementation details: I did suspect you had a very good reason not to force a perfect match in character placement. But if you can find some time to code this option, I would be really grateful. And your idea sounds precisely spot on: an option to force a perfect match in character placement, plus a slider to tweak the font-width algorithm. Thanks again.
  12. Hi WildBill, I have been comparing screenshots, both on a Windows 2000 and on a Windows XP system. The idea was to try and identify 1) any remaining, visual gap between ClearType and SmoothText, and/or 2) any unexpected difference introduced by SmoothText when compared with standard, no anti-aliased text. What I found, for both comparisons, is the same result: SmoothText doesn't seem to preserve the overall width of the text. This problem affects different fonts and different sizes in different ways, but it always manifests itself either as a increased/reduced character spacing or an increased/reduced character width (when compared with the same text, rendered with ClearType or no anti-aliasing at all). To some extent, this issue seems to depend on the Energy slider setting, but not in an obvious, consistant way: e.g. on the same page, with SmoothText at Energy 6, some text gets slightly expanded in overall width (as SmoothText is turned on), while some other text gets compressed. The problem occurs for all the settings of the Energy slider. On the contrary, ClearType doesn't introduce any difference in inter-character spacing or character width. So, as all we're comparing here is width, the comparison is equally valid on Windows 2000 (SmoothText vs. no anti-aliasing) or Windows XP (SmoothText vs. Cleartype). I have attached two screenshots, the first with ClearType, the second with SmoothText 0.7.2 at Energy6/Gamma1.40 (both on Windows XP). You can look for yourself at the site I have used as reference, BBC News (tested with Firefox 3.0.6 and Internet Explorer 6.0): http://news.bbc.co.uk/ (Note: At first I thought that this problem had been introduced with the Energy slider, I as have older screenshots of the same website, captured with SmoothText 0.6.1, which were perfetct, width-wise. But then I looked at other screenshots, same website and still with 0.6.1, but on a different system, and the problem was already there). Another good test website is eBay, as it uses a lot of Arial and Times New Roman in different sizes and styles (bold, italic, etc.).
  13. Excellent work with the Gamma correction slider. It mimics ClearType Tuner's Contrast setting very closely. I haven't had much time to choose a proper Energy/Gamma setting yet, but it's amazing how finely SmoothText can be tuned with these two sliders. By the way, very interesting material on Gamma correction: thanks for sharing.
  14. Thanks a lot for implementing the feature this quickly. I have downloaded version 0.7.1 and I really like the results. From a few basic tests I have run, SmoothText with a 30% setting for the energy distribution slider (0% being the leftmost setting) looks virtually identical to ClearType. Note: I used a PC running Windows XP to get results as accurate as possible, with Firefox 3.0.6, Internet Explorer 6.0, Word 2003, and 10pt Verdana for all the tests. Please let me know if there are other specific tests worth running and/or if you want screenshots of some sort.
  15. Hi WildBill, First things first, I have only been using SmoothText for a few weeks, but I find it really great. I would like to suggest a new feature, for your consideration. Would it be possible to add a slider, to customise the "level" of anti-aliasing rendered by SmoothText? The slider would have 0 on one end (no anti-aliasing at all) and 100% at the other end, corresponding to a really "thick" and/or smoothed font rendering. 50% could be the current behaviour. Not knowing the details of the present rendering algorithm, I would imagine that the slider could tune the width of the energy distribution curve around the central (sub-)pixel. In practice, this would work in a way somewhat similar to the Advanced tab in Microsoft's ClearType Tuner PowerToy (Windows XP only): http://www.microsoft.com/typography/ClearTypePowerToy.mspx Or, to give an example which runs on most OSs, this would work a bit like the Tune function in this freeware little demo: http://www.grc.com/freeandclear.htm FYI - I'm suggesting this new feature as your very first post got me thinking: "It's not quite ClearType™, but...". Assuming it's feasible, this could virtually close the gap.

  • Create New...