Post a Comment
The Javascript engine written to take advantage of any potential performance improvements that could be yielded. Who said it was a magic wand to speed things up when you're too intellectually bankrupt to read the damn message from start to finish within complete context.
Opera's support for plugin's isnt' that bad. flash has been problematic recently (shockwave specifically) but Unity (I assume you mean http://unity3d.com/webplayer/ ) and many others (quick time also isnt perfect, but its not great on the windows playform either).
They already had a button for the menu. It just wasn't default. Just click File->Show Menu Bar.
I like the removal of modal dialogs, but I'm not sure about the implementation of it. As a regular Opera user, I have been annoyed by dialogs preventing me from switching tabs, but you make a good point about phishing.
Also, check out http://my.opera.com/ODIN/blog/show.dml/5513451#comment13791981. It looks like HTML5 video will be present in 10.5.
Actually, that didn't make it into this build! Neither did video (which they initially planned according to Twitter comments). So imagine the speed when HW acceleration is turned on!
Will Vega acceleration be across the board? will that mean that Opera will end up using Direct2D/DirectWrite? hopefully this will be the version of Opera that will pull me over - but then again, I have been spoilt by process separation and plugin isolation which chrome brings to the tale - that is the one and pretty much only reason I use Chrome
Well, Vega is their graphics library and it powers both webpages and the UI, but that doesn't mean that it's hardware accelerated. Currently it's software based and they still have about 4 times faster speed. According to some of the developers' blogs it is designed so that it is trvial to add a hardware acceleration backend and they have Direct3D and OpenGL backends, but they don't seem to be needing them
sorry i am typing on mobile phone.
i stopped reading osnews for a while because the review for opera 10 was bias and unfair. (this is coming from a fan of firefox more than opera). the desinger who designed the opera and firefox buttons is the same person. those buttons style was used by opera first, so its firefox that is copying opera (among many other things). although i do have to admit Opera need to listen to customers more, fixing bugs and replying to customers request, this will gain them more users. if they want to compete with firefox, they'll need an addon system (not the limiting widgets.)
there are still more things firefox need to copy off from opera to improve, such as the bookmarking system. adding and maintaining bookmarks in firefox is a pain (the very little window that pops up when you press ctrl+d). it can't even sort them in alphabet order for christ sakes. i hear there are some people who uses "tab mix plus" to have 500 tabs opened at the same time in firefox. if they could maintain bookmarks as easy like opera they wouldn't had to require 500 tabs.
the opera ui provides maximum freedom, which is what osnews criticize about. people prefer to be limited of functions than to have too much. so in a way, it seem to backfired for people who are used to the firefox ui which is fixed solid and left-to-right oriented (unless with addons), which really came from internet explorer interface.
with that said, i am so spoiled with firefox addons that i will not likely switch back to it soon, unless there is noscript for opera (i know there are some, but doesn't work too well).
Edited 2009-12-22 22:43 UTC
the default codec still hasn't been decided.
As for speed I think all these javascript benchmarks are rather silly. It isn't as if when you visit websites your cpu is choking on javascript. If your cpu chokes on anything it is Flash. It reminds of video card benchmarks that tout a gpu as 'the best' for having superior gpl performance when the games that challenge the gpu are written in directx.
its not like they don't have time. HTML5 isn't scheduled to be a proposed recommendation until 2022 http://blogs.techrepublic.com.com/programming-and-development/?p=71...
browsers can support whatever they like, the problem is that the standard itself is going to be up in the air until then. Considering that there is currently no free implementation that is good enough for google to consider using with youtube, I wouldn't be suprised if they just wait another 5-10 years to hammer out that detail.
Just because it isn't a final recommendation doesn't mean it's up in the air. Browsers will work on aligning implementations all along, and the more mature the spec becomes, the fewer changes, even if the final spec is years away.
You need to educate yourself about how the process works before spouting nonsense.
So maybe you didn't get what I was saying; this is one thing where the different vendors have different requirements and there is no technical solution that currently exists that meets everyones needs. This is one of those details that aren't going to get aligned for awhile, and it doesn't really have to be because we are nowhere near deadlines for it.
It takes a long time to do things properly. It would have been quicker if "someone" (hello, Microsoft and Apple!) didn't keep sabotaging the process.
Also, just because the final recommendation isn't out until 202x doesn't mean that the spec won't be usable before that. The more it matures the better the browser support will be, and that will happen many years before the final spec is out.
There is no such thing as a default codec. And there shouldn’t be. There should be consensus, yes, but no one mandated codec.
When the image tag was created the spec could have mandated the XBM format, since that was common at the time.
Mandating a codec now will be just as out of date and impractical as only allowing XBM files for image tags in many year’s time.
What's the Chrome way, though? Hidden menus are the default in Vista/Windows 7. Tabs above the address bar is something Opera did ages ago. Hiding the title bar perhaps? That's the only new thing Chrome really did I guess.
...Visited a few sites. Not too shabby! I do like how it integrates with W7. Looks good. Some visual glitches (as was mentioned) while browsing but if it is fairly decent now, I can imagine it will be a fine piece of software when it is released. I should have gone and tried it on those [work] web apps that never worked with opera in the past. Maybe later (I do have to submit my time tracking for the week).
I really don't think it matters that Opera doesn't support HTML5 video yet. Opera's market share is extremely small, especially compared to IE. The only sites that will start using the video tag before IE supports it are sites with a target audience of more technically inclined people and that don't mind telling IE users to switch browsers. If they are okay with leaving out IE users, I'm sure leaving out Opera users would be okay too. And I'm sure that if IE did implement it, Opera support would soon follow. So it really doesn't matter.
Also, I find it rather ironic that there is another unrelated vector graphics library with the name of vega (it is a OpenVG state tracker for Gallium3D). Coincidence? I think not!
It’s not about leaving other browser users out-- http://camendesign.com/code/video_for_everybody It’s about providing a _better_ video experience for the websites that do use HTML5.
Carakan brings some improvements that no tech site mentions and that are relatively buried in Opera's site.
With Carakan each tab gets its own heap for memory so a common operation of javascript (garbage collection) becomes a lot faster. Basicly the speed benefits of Chrome's process separation without the stability.
More importantly with Carakan the engine cahces compiled code. That means that the second time the engine encounters the same code it's going to execute much faster. Obviously this is something that has a huge impact in real world for when you are loading multiple pages of the same site, when you are reading different news stories on a site etc.
Linux/UNIX users might want to give this a scan through whilst they await versions of Opera 10.50 for their own platforms:
http://my.opera.com/ruario/blog/unix-10-50-evenes-work-in-progress
Edited 2009-12-23 07:26 UTC
Yes. So?
Maybe you should read what the developers are saying instead of being an ass about it:
http://my.opera.com/ruario/blog/unix-10-50-evenes-work-in-progress#...
I've used opera for years now and I will most probably continue to do so. A major reason is that opera only uses 1 window, others are the speed dial and opera unite.
I'm typing this comment in the new pre-alpha on a virtual xp and it works great. I hope the linux version is just as good. Thanks to Opera Unite I could try my favorite sites very quickly. (when will others follow?)
Dialogs in page is just a very good idea, why stop the whole browsing experience for one page.
I've read about extensions in the comments. I won't say no to them, but the way they are implemented now in other browsers is not the way to go for me. I tried to use them with firefox, but finding good ones for me was a nightmare.
I've used opera for years now and I will most probably continue to do so. A major reason is that opera only uses 1 window, others are the speed dial and opera unite.
I'm typing this comment in the new pre-alpha on a virtual xp and it works great. I hope the linux version is just as good. Thanks to Opera Unite I could try my favorite sites very quickly. (when will others follow?)
Dialogs in page is just a very good idea, why stop the whole browsing experience for one page. The effect on security just depends on how it is done under the hood.
I've read about extensions in the comments. I won't say no to them, but the way they are implemented now in other browsers is not the way to go for me. I tried to use them with firefox, but finding good ones for me was a nightmare.
Mozilla folks think it's the opposite: http://design-noir.de/log/2009/12/opera-firefox-ui/
Mozilla folks think it's the opposite: http://design-noir.de/log/2009/12/opera-firefox-ui/ "
That's because the Mozilla people are evidently retarded. A leaked version of the Opera pre-alpha was released two days before the Firefox UI mockups were posted. So if anything, Mozilla stole from Opera. Again.
And the Mozilla mockup goes back to December. http://design-noir.de/log/2009/12/opera-firefox-ui/#comment-14680
What matters more than "who killed who" is consensus. The more apps that adopt the idea, the better.
The Mozilla mockup did not appear until two days after Opera 10.5 was leaked. They may have been presented at a Mozilla meeting, but I seriously doubt that people from Opera are attending internal Mozilla meetings!
Never mind the fact that the button is already in Opera 10.0, only disabled by default.
Some cool stuff in this build, but the first thing I notice is that MDI browsing is completely broken. I really hope that this is something that'll be fixed in the release version; to me MDI's flexibility constitutes about 90% of the reason why I prefer its UI to other browsers.
Can I just say how much I loathe and despise pointlessly non-standard GUIs - with their oddly shaped windows, buttons that are inconsistent with anything else, and more often than not reduced functionality and damaged usability. That’s one of the reasons I took an instant dislike to Chrome, use Foobar 2000 and VLC rather than fancy skinned media players, and usually breathe a sigh of relief when I turn on Opera’s Windows Native skin.
I hate to see Opera breaking its best features and going in that non-standard direction.
I just ran Futuremark Peacekeeper to compare the latest browsers on my MacbookPro 4,1 (Snow Leopard 10.6.2) and here are the results:
Safari 4.0.4 (Build r52531) - 4331
Chrome 4.0.282.0 (Build 35277) - 3390
Opera 10.50 (Pre Alpha) - 2350
Minefield 3.7a1pre (Gecko/20091225)- 2274
Firefox 3.5.6 - 1499
Interesting results methinks!
I just ran Peacekeeper in the Windows version of Opera 10.5 using Crossover 8 on the same machine.
Safari 4.0.4 (Build r52531) - 4331
Chrome 4.0.282.0 (Build 35277) - 3390
Opera 10.50 Win32 (Pre Alpha) - 2580
Opera 10.50 OSX (Pre Alpha) - 2350
Minefield 3.7a1pre (Gecko/20091225)- 2274
Firefox 3.5.6 - 1499
Slight improvement over the OSX native version but still way short of Chromium and even more so of Safari.



