Presto engine

[sigmark]

sigmark last edited by

Originally posted by RoadHazard:

Moving to Webkit/Blink was a good idea, it will ensure great compatibility and allow the Opera devs to focus on browser features rather than engine development. They just need to get the features back in there.

aye, and the focus on the browser features bereft us of all the features it had. now it is no better than chrome

[berng]

berng last edited by

Originally posted by Sigmark:

haavard:
you dont mean it do you? chrome opens.every window as a standalone process and in sum it can take up way above 1/2 gigabyte of RAM.

With opera as it was I never.managed to get over 1/4 gigabyte with 40 windows open simultaneously.

at least where the system resources are concerned

Do you have an issue with page faulting? If not it doesn't matter. If the ram is there its to be used, provided the coding is efficient. Unused ram is wasted ram.

I once changed a program that read in buffers of 50 records to a buffer, allowing 20 buffers. Before it read unbuffered one record at a time. It was our nightly input file conversion. A buffer would be read, the 50 records processed in ram and when done the changed buffer would be written to disk. While one buffer was processed, other buffers would be read from the disk drives. This means when I was done with a buffer and the new buffer was being written to disk, a new buffer was being processed in ram. The nightly batch job went from 3 hours to 5 minutes. Then someone criticized because I was using so much "ram." Didn't matter. The ram was used well and operations was ecstatic because that 3 hours batch run which held dozens of other jobs up was reduced to 5 min.

If you can, try to process your data in ram. Its fastest.

[vikingen]

vikingen last edited by

Originally posted by haavard:

Originally posted by rs79:

What exactly is gained by the move to webkit?

A major boost in site compatibility, and freeing up resources to do interesting things instead of constantly having to fight the web.

Originally posted by Vikingen:

I think Presto have less and less compatibility problems.

Unfortunately, this is not the case. It became more and more of an uphill battle to keep Presto compatible with the web.

I didn't know that because my personal experience is a bit different.

Why is it hard to keep Presto compatible? What happened to the standards? There was a time when much of the web was coded to work with IE6 but that was way back in the past.

[frenzie]

frenzie last edited by

Originally posted by Vikingen:

Why is it hard to keep Presto compatible? What happened to the standards? There was a time when much of the web was coded to work with IE6 but that was way back in the past.

Webkit became the new IE6. See e.g. http://www.opera.com/docs/specs/presto2.12/css/aliases/

Blink actually promises not to do that anymore: http://www.chromium.org/blink/developer-faq#TOC-What-s-stopping-Chrome-from-shipping-proprietary-features-

(Then again, they shipped Dart…)

[sigmark]

sigmark last edited by

Originally posted by BernG:

Originally posted by Sigmark:

haavard:
you dont mean it do you? chrome opens.every window as a standalone process and in sum it can take up way above 1/2 gigabyte of RAM.

With opera as it was I never.managed to get over 1/4 gigabyte with 40 windows open simultaneously.

at least where the system resources are concerned

Do you have an issue with page faulting? If not it doesn't matter. If the ram is there its to be used, provided the coding is efficient. Unused ram is wasted ram.

If you can, try to process your data in ram. Its fastest.

youre carrying wood into forrest
If chrome takes ram it forces the system to use swap file, which loads up the drive and slows all operations. its the principle of the thing! I have issue of having sofware junk up my pc when i dont ( didnt) have to. Opera kept the resources lean and did not waste them. wasted ram is as wasted as unused one, but since the browser is only one of the activities simultaneously running, its the share of the cake it fills that annoys me. and we are told resources are being freed by moving to big brother googles chromoid clone

[sgunhouse]

sgunhouse Moderator Volunteer last edited by

There is a command-line option to force Blink into single-process mode, but I've never used it and so can't say how much performance you lose that way.

[greekonsun]

greekonsun last edited by

Originally posted by Vikingen:

Why is it hard to keep Presto compatible? What happened to the standards?

I don't think the switch happened b'coz of compatibility problems
in fact Presto was always best at W3C standards

it was just the fact that webkit became dominant, no matter how shitty it is
so certain pages were either made just for sakes of webkit and started to block Opera completely

try to go to icloud.com with "older opera", you will get "Unsupported browser" error crap and be blocked
now hack your output to lets say Firefox 25, behold the miracle, no error message and webpage works perfectly fine

it is just whole browser wars politics fiasco crap
so normally its easier for Opera devs to jump onto webkit train and do lazy approach instead continuing with their unique engine that "nobody wants"

[vikingen]

vikingen last edited by

So why don't everybody else switch to Presto, perhaps because Opera didn't want to give it away for free? Presto could have become open source. Or maybe it is already? I don't know.

[blackbird71]

blackbird71 last edited by

Originally posted by Vikingen:

So why don't everybody else switch to Presto, perhaps because Opera didn't want to give it away for free? Presto could have become open source. Or maybe it is already? I don't know.

It's not open source, and Opera spent a lot of years and money refining it into what it became. Moreover, it is still paid-licensed to vendors for use in a few non-Opera products. I believe Presto was built during an era when Opera was corporately focused on providing user features and configurability, so Presto incorporated a number of things that made Opera's user interface work better in those areas. As such, and being closed code, other browser makers chose to develop their own code instead, focusing on different target markets. And, being bigger outfits and better at marketing their products in a variety of ways, they came to dominate the non-IE marketplace. At this point, that's all simply history. Presto Opera is over, other than 12.16 which is merely on temporary life-support until Opera pulls that plug.

Some of us as users may not like the way things have worked out, but they are what they are; and we are mere users of a free browser that is being totally redesigned. We need to get over dreaming that Opera is going to somehow change its mind and revert its entire product line back to Presto... after a year or more of pursuing the costly conversion to Blink, it just isn't going to happen, period! End of story! Either we find other browsers to use or we hang in there with Old Opera if and until New Opera becomes palatable to us.

[vikingen]

vikingen last edited by

Originally posted by blackbird71:

Originally posted by Vikingen:

So why don't everybody else switch to Presto, perhaps because Opera didn't want to give it away for free? Presto could have become open source. Or maybe it is already? I don't know.

It's not open source, and Opera spent a lot of years and money refining it into what it became. Moreover, it is still paid-licensed to vendors for use in a few non-Opera products. I believe Presto was built during an era when Opera was corporately focused on providing user features and configurability, so Presto incorporated a number of things that made Opera's user interface work better in those areas. As such, and being closed code, other browser makers chose to develop their own code instead, focusing on different target markets.

I'm thinking that it would be better to give away the source code instead of burying it. Opera Software lose either way.

[stng]

stng last edited by

Originally posted by Vikingen:

Originally posted by blackbird71:

Originally posted by Vikingen:

So why don't everybody else switch to Presto, perhaps because Opera didn't want to give it away for free? Presto could have become open source. Or maybe it is already? I don't know.

It's not open source, and Opera spent a lot of years and money refining it into what it became. Moreover, it is still paid-licensed to vendors for use in a few non-Opera products. I believe Presto was built during an era when Opera was corporately focused on providing user features and configurability, so Presto incorporated a number of things that made Opera's user interface work better in those areas. As such, and being closed code, other browser makers chose to develop their own code instead, focusing on different target markets.

I'm thinking that it would be better to give away the source code instead of burying it. Opera Software lose either way.

Well, according to their PR team member:

1. Licensing work costs too much money and man-hours. Don't forget about patents which were used in Presto's code.
2. They probably won't want to share their intellectual property with everyone.
3. Presto is still here on a various types of devices (e.g.TVs)

So it's likely that Presto source code will be buried. :furious: :rip:

[Deleted User]

Deleted User last edited by

Originally posted by STNG:

...
3. Presto is still here on a various types of devices (e.g.TVs)
...

I've been to Opera's booth at the IBC-2013 show, they were presenting some smart TVs.
According to the guy there - it is not Presto any more. He said the TV team has had to switch to chrome too.

If they bury the Presto/Caracan - it will be the silliest thing to do ever!

[blackbird71]

blackbird71 last edited by

Originally posted by Vikingen:

Originally posted by blackbird71:

Originally posted by Vikingen:

So why don't everybody else switch to Presto, perhaps because Opera didn't want to give it away for free? Presto could have become open source. Or maybe it is already? I don't know.

It's not open source, and Opera spent a lot of years and money refining it into what it became. Moreover, it is still paid-licensed to vendors for use in a few non-Opera products. I believe Presto was built during an era when Opera was corporately focused on providing user features and configurability, so Presto incorporated a number of things that made Opera's user interface work better in those areas. As such, and being closed code, other browser makers chose to develop their own code instead, focusing on different target markets.

I'm thinking that it would be better to give away the source code instead of burying it. Opera Software lose either way.

Think it all the way through. If Opera were to give away Presto to be open source, what would it accomplish? Gain the undying gratitude of what it currently believes is a small component of the market's user base? Perhaps. But for sure and for certain, it will cause to come into existence new browser (and other kinds of software) designs from a variety of directions. And, whether good designs or not, those will act to compete in the marketplace for the marketshare that Opera is already trying to increase with its Blink browsers and remaining Presto products. Apart from a sudden burst of altruism, why would any company freely release code it had spent years and lots of money to develop with the only real prospect of that being increased competition for that company's remaining products? This is one of the key reasons Microsoft has never released any of its old operating systems into the public domain - not even DOS.

[frenzie]

frenzie last edited by

Originally posted by blackbird71:

Apart from a sudden burst of altruism, why would any company freely release code it had spent years and lots of money to develop with the only real prospect of that being increased competition for that company's remaining products?

id Tech 2: released 1997, open-sourced 2001
id Tech 3: released 1999, open-sourced 2005
id Tech 4: released 2004, open-sourced 2011

One can hope.

[vikingen]

vikingen last edited by

Originally posted by blackbird71:

Originally posted by Vikingen:

I'm thinking that it would be better to give away the source code instead of burying it. Opera Software lose either way.

Think it all the way through. If Opera were to give away Presto to be open source, what would it accomplish? Gain the undying gratitude of what it currently believes is a small component of the market's user base? Perhaps. But for sure and for certain, it will cause to come into existence new browser (and other kinds of software) designs from a variety of directions. And, whether good designs or not, those will act to compete in the marketplace for the marketshare that Opera is already trying to increase with its Blink browsers and remaining Presto products. Apart from a sudden burst of altruism, why would any company freely release code it had spent years and lots of money to develop with the only real prospect of that being increased competition for that company's remaining products? This is one of the key reasons Microsoft has never released any of its old operating systems into the public domain - not even DOS.

Opera has to compete with other browsers regardless. If other companies moved to Presto, at least Opera would compete on a platform they know. Perhaps competing on the Presto platform wouldn't be an advantage for Opera, but at least the best platform would win.

Microsoft and DOS is different. Microsoft never had the choice between releasing DOS or converting. Microsoft have no reason to release DOS because DOS is obsolete. And Microsoft would lose money if they released old Windows versions because some people would use them instead of buying win 8. Opera Software don't offer a new browser for money so they would not lose money by releasing Presto. Presto is not technically obsolete, it's going obsolete only because no one will continue to support it.

By the way, are Blink and WebKit related?

[frenzie]

frenzie last edited by

Originally posted by Vikingen:

By the way, are Blink and WebKit related?

For pretty much all intents and purposes, Blink is still a variety of Webkit. See http://en.wikipedia.org/wiki/Blink_(layout_engine)

[greekonsun]

greekonsun last edited by

Originally posted by booBot:

If they bury the Presto/Caracan - it will be the silliest thing to do ever!

more like dumbest thing

// to others above
at least Open sourcing would allow people to have branched off version of Opera
just like other browsers based on Mozilla branched off Firefox which is another disaster now

if they gonna throw it away, why waste it when can be saved and blossom in something great ?
unless Google payed them off somehow to do it on purpose

[blackbird71]

blackbird71 last edited by

Originally posted by Vikingen:

...Opera has to compete with other browsers regardless. If other companies moved to Presto, at least Opera would compete on a platform they know. Perhaps competing on the Presto platform wouldn't be an advantage for Opera, but at least the best platform would win.

Microsoft and DOS is different. Microsoft never had the choice between releasing DOS or converting. Microsoft have no reason to release DOS because DOS is obsolete. And Microsoft would lose money if they released old Windows versions because some people would use them instead of buying win 8. Opera Software don't offer a new browser for money so they would not lose money by releasing Presto. Presto is not technically obsolete, it's going obsolete only because no one will continue to support it. ...

You're mixing two different issues: whether Opera should release Presto to open source, and whether Opera should continue basing its browsers on Presto.

If Opera attempted to somehow make it more economical to maintain Presto as a browser base by releasing it to open source, Opera would then have to sit back and wait for "other companies to move to Presto" to gain anything... which could take eons (in "Internet-years") to achieve. But with Opera remaining in the Presto game, a major incentive for others to enter that game would evaporate, since they would be starting "behind the curve" compared with Opera's Presto experience. In the meantime, Opera would lose any revenue they receive from licensing Presto to customers for other apps. Moreover, to release Presto into public domain with any hope of gaining adherents that could contribute meaningfully, Opera would likely incur significant overhead costs (documentation especially). Managing open-source code (and all the ensuing forks) has its own costs, downsides, and compatibility nightmares.

Microsoft and DOS are not different, in terms of how a company that expends resources to develop code views it. Companies, in writing and testing solid code at great expense, develop algorithms and techniques that are proprietary to them. Some of that resides in the limitations of the codebase being written, but some of it lives on in the general way a given technique unfolds and becomes grafted-in to a new codebase, should the codebase be changed. Moreover, codebase transitions are rarely "instantaneous"... Windows, after adopting the GUI interface, used DOS elements for many years embedded within. This was just one aspect of proprietary techniques having a life well beyond their original paradigm. Companies understand this quite well, at least companies that survive for very long.

Opera makes money from its browsers in several ways. By licensing pre-load versions to phone makers, by renting out preferential positioning (for search engines, Discover content, SpeedDial defaults, etc), and by licensing use of its code. The greater the market share of its browsers, the higher the rates it can charge and the greater the usage-related payments. So anything Opera might do to increase competitor browsers will act to decrease their own market share and revenue, regardless of whether or not the browser is free. It's how the free-browser market actually works. In some cases, companies that have gigantic business bases in other arenas may offer free browsers for publicity purposes or to enhance their other business areas, but Opera has never been one of those companies... it has to deal with the realities of its own position in the browser markets.

In any case, the reality is that Opera has moved away from Presto in its browsers. That is an essentially irreversible business decision. Whether it ever releases Presto to open source obviously remains to be seen, but I seriously doubt it will ever happen, simply for the economic reasons I've noted.

[blackbird71]

blackbird71 last edited by

Originally posted by greekonsun:

...at least Open sourcing would allow people to have branched off version of Opera
just like other browsers based on Mozilla branched off Firefox which is another disaster now

if they gonna throw it away, why waste it when can be saved and blossom in something great ?
unless Google payed them off somehow to do it on purpose

Unlike the original Mozilla Foundation, Opera is not in a business of developing code and giving it away free for the world to use. As with all private-enterprise, for-profit companies, it's in the business of making money for its investors (without which there would be no incentive for investor folk to have risked their money in the first place). As such, Opera has to look at things from the perspective of what decisions and actions will enhance its economic and financial positions and offer the best chances of achieving gains for its stockholders.

We may disagree with choices Opera makes in carrying out its business objectives, and we may believe we have very good reasons for disagreeing, but we have not been charged by Opera's owners (its stockholders) with performing financial diligence with their investment. That battle, if any, lies between Opera's stockholders and Opera's business managers. Put simply, it is none of our business, unless we are stockholders.

[vikingen]

vikingen last edited by

Originally posted by blackbird71:

Originally posted by Vikingen:

......

........

Thanks for replying, I get it now. You understand it better than me it seems.