–Gadget, too, also likes fantasy
–Gadget, too, also likes fantasy
This was a TiDi-trap, as I mentioned. I think, the core Imperium leaders thought that the North leaders have very overestimated position over EVE battles and skills. They thought the North will come with all their Supers to show their power like at Auga Keepstar battle and during the final battle of Tribute campaign.
Dear Gadget, I wish you to become a Clever Gadget also and check everything not only in EVE, but in real life also.
There are a way too many historical events which are covered by lie, secret protocols and other cases due to political, propaganda, domination and other reasons.
If you want to know the truth about North Korea just don’t watch only TV or don’t read only MARVEL magazines. You have to ‘dive deeper’, you have to read a lot of historical, journalist texts from all nearby countries: Russia, China, South Korea, Japan and North Korea of course. You have to travel there and speak with those people, from different social groups.
If to be more accurate, the first steam powered ship officially was launched in 1807. To know more about the steamboats you have to read not from official political protocols of big heroic battles, but you have to study the life of Robert Fulton (here’s a simple biography).
Robert Fulton (November 14, 1765 – February 24, 1815) was an American engineer and inventor who is widely credited with developing a commercially successful steamboat called The North River Steamboat of Clermonts. That steamboat went with passengers from New York City to Albany and back again, a round trip of 300 miles, in 62 hours in 1807. In 1800, he was commissioned by Napoleon Bonaparte to attempt to design Nautilus, which was the first practical submarine in history. He is also credited with inventing some of the world’s earliest naval torpedoes for use by the British Royal Navy.
Fulton became interested in steam engines and the idea of steamboats in 1777 when he was around age 12 and visited state delegate William Henry of Lancaster, Pennsylvania, who himself had earlier learned about inventor James Watt and his Watt steam engine on a visit to England.
Fulton came to England in 1786, carrying several letters of introduction to Americans abroad from the individuals he had met in Philadelphia. He had already corresponded with Benjamin West, and West took Fulton into his home, where Fulton lived for several years. Fulton gained many commissions painting portraits and landscapes, which allowed him to support himself, but he continually experimented with mechanical inventions.
As early as 1793, Fulton proposed plans for steam-powered vessels to both the United States and British governments, and in England he met the Duke of Bridgewater Francis Egerton, whose canal, the first to be constructed in Britain, was being used for trials of a steam tug. Fulton became very enthusiastic about the canals and in 1796 wrote a treatise on canal construction, suggesting improvements to locks and other features. Working for the Duke of Bridgewater between 1796 and 1799, he had a boat constructed in the Duke’s timber yard, under the supervision of Benjamin Powell. After installation of the machinery supplied by the engineers Bateman and Sherratt of Salford, the boat was duly christened “Bonaparte” in honour of Fulton having served under Napoleon. After expensive trials, because of the configuration of the design, it was feared the paddles may damage the clay lining of the canal and the experiment was eventually abandoned. In 1801 the Duke, impressed by Charlotte Dundas constructed by William Symington, decided to order eight of such vessels for his canal, but when he died in 1803, the order was cancelled. Symington had successfully tried steamboats in 1788, and it seems probable that Fulton was aware of these developments.
In 1804, Fulton switched allegiance and moved to England, where he was commissioned by the Prime Minister William Pitt the Younger to build a range of weapons for use by the Royal Navy during Napoleon’s invasion scares. Among his inventions were the world’s first modern naval “torpedoes” (modern “mines”), which were tested, along with several other of his inventions, during the 1804 Raid on Boulogne, but met with limited success. Although he continued to develop his inventions with the British until 1806, the decisive naval victory by Admiral Horatio Nelson at the 1805 Battle of Trafalgar greatly reduced the risk of French invasion, and Fulton found himself being increasingly ignored.
Based on this example as an ‘assumption’, I can ‘affirm’ that the North won the Keepstar batlle not because some miserable TiDi, but because they were clever, used good tactics to mitigate dps and did all like heroes. This is a political side of that Keepstar battle.
Get up, Prince of Troy. Get up. I won’t let a stone take my glory.
NOPE. Don’t care about your silly space battle.
I was working, so I didn’t get to see the (not)stream.
I wish the victors and the defeated well.
I hope that technology (or decent enforced policies) can be used to mitigate the stagnation.
There will be a next time.
With that, I return to important subjects:
Which of Nelson’s ships had steam power?
That is what you posited; that is what I quickly denied.
Your own source (Wikpedia… Really? Professional pride hurt ) says nothing about steam powered ships in Nelson’s fleet. There’s even a link to the Order of Battle.
(J/K about the Wikpedia - One has to use whatever is available)
It was like 5000 people tried to fight side by side in a huge pool filled with honey. You barely can do anything.
Shameless self-promotion of a potential fix to the whole TiDi mess.
I’m only linking it here because TiDi is a frequent topic in the thread.
It’s true and the TiDi issue in EVE persist since Fountain War, based on those official videos and articles. Since there in every big battle the TiDi played huge role, but in this Cloud Ring system it was crucial.
To solve the tech issue isn’t easy. It requires huge changes in EVE server code, used DB-servers and user client code. Powerful and fast servers are expensive and can cost much more than entire CCP company. Who can just pay for subscription. The user clients should support CUDA computation technologies.
CUDA® is a parallel computing platform and programming model developed by NVIDIA for general computing on graphical processing units (GPUs). With CUDA, developers are able to dramatically speed up computing applications by harnessing the power of GPUs.
In GPU-accelerated applications, the sequential part of the workload runs on the CPU – which is optimized for single-threaded performance – while the compute intensive portion of the application runs on thousands of GPU cores in parallel. When using CUDA, developers program in popular languages such as C, C++, Fortran, Python and MATLAB and express parallelism through extensions in the form of a few basic keywords.
The CUDA Toolkit from NVIDIA provides everything you need to develop GPU-accelerated applications. The CUDA Toolkit includes GPU-accelerated libraries, a compiler, development tools and the CUDA runtime.
The CUDA problem is: every client machine should have NVidia card and that card should be modern like NVidia GTX 1050, 1060, 1070, 1080 and the game client should completely compute visual objects in CUDA. The Radeon version has a different lite version based on OpenCL.
How CUDA works:
Also, the interface must contain some kind of AI to auto disable/enable game settings, in similar game scenarios: D-Scan, Local list, other slow UI, visual effects and animation, to display tables like lists (I have huge experience by developing DB and their UIs. The performance can be greatly improved by buffering them and by disabling table grid lines), and etc and etc.
I’m certain CCP would have at least considered CUDA by now… I’m not sure how many of the calculations would really run faster on GPUs though. A lot of the load would be sanity checks. “Is the module off cooldown, what is the current transversal speed of the target”.
I agree CUDA could definitely help with the actual calculations, which likely represent a big part of the load, but another big part would not be able to play to a GPU’s strengths.
CUDA is the future of massive client side graphic computations and visualization. In the example above there are 32700 objects rendered under 160fps based on PhysX interactions between them. I watched many CUDA samples and they look impressive. Sadly, current CUDA is used in science mainly to emulate, for example, molecular or biological reactions.
[updated] Here, I found that impressive sample I saw long time ago
Moreover, CCP should create a kind of exchange protocol to send and receive data between server and clients in different channels (if it’s not yet implemented), like current ATM works, to mitigate delayed user commands. And the data in protocols should be well compressed like is JSON, for example.
TiDi has absolutely nothing to do with client-side lag. Yes I agree that’s also a problem, but that’s a problem that can be assisted by turning down your graphics, turning off brackets, etc.
The graphical load on the client is absolutely important, but it is unrelated to server lag. The server doesn’t render anything.
TiDi is an artificial server latency to mitigate the crash or shutdown. It is a sum of downgraded grid thread cycle time + a computation time + DB I/O time (read/write to HDD). Total latency is TiDi + pack and process data in routers + packets travel time to clients.
Moreover, CUDA can be used for server-side computations only without 3D visualization, but this usage of the technology isn’t popular yet and well known. The tech is new (from 2008-2010) and for true server parallel computation should be used some tricks. I’m not familiar with these ‘aspects’ yet.
Actually, for the most part we didn’t use drones and that was one of the reasons why. You can’t use sub caps because all the defenders would have to do is use a GTFO module and with the ECM burst fleet that they brought ships would have to re-establish locks every shot. This made fighters the best option.
Like I said…
For example, we all know the type of guy who will be smashing a hotkey as hard as he can while the module is cycling, over and over again. I have to assume the client will send at least some of those through to account for potential client lag. Each of those module availability checks incurs a check that the server must handle. Granted that is small, but amplify it across thousands of players, in a situation where the processors are already very close to thrashing, and it probably matters a lot.
Or d-scan. No real computations required (just a quick subtraction assuming that’s done on the server at all, it would likely be more efficient to return xyz and let the client figure it out) however it has to query the position of every object within 15 AU. 5000 objects to query, results for 5000 objects to return. That adds up quick.
CUDA would certainly help for the actual calculations, such as damage from a weapon firing, but those are only part of the load. Not saying it won’t help, just saying CUDA isn’t a silver bullet.
I like Grath Telkin a lot, because he just speaks his mind and he has passion. On the face of it he is right from the perception of PL/NC/GOTG, there was no strategic value to that Keepstar in the sense of holding meaningful assets and being unimportant.
Listening between the lines to things it was evident that it was a test for the Imperium, to see what they could do, they wanted to try different tactics to take out a Keepstar when defended by a large force. Their tactics would have worked if they could have reloaded their fighter tubes. What this in reality meant is that if you defend a Keepstar you have to quite simply turn it into a TiDi fest so nothing works. What I did learn from that is that outside of the main TZ’s a super blob without excessive server load will be able to kill a defended Keepstar. I was there at the destruction of the GOTG Keepstar and the DRF dropped Dreads to keep the damage cap up, I was shooting GOTG fighters in a Jackdaw.
Goons are not bothered so much by the strategic importance of that Keepstar, they are coming at it from a different angle, this is a typical Goon thing which is to do it because they can and in doing that they install a certain amount of fear and fear is control. The thing is that they can destroy Keepstars and at some point you will see it happen.
The Goons have built a massive amount of Supers and Titans, it does not really matter where they head in terms of attacking Keepstars, but the simple fact is that they will, they could for a start go after Test, it would not surprise me at all to see the Keepstar which my two mains are currently sitting in to be the one they go for, though that splat between Aryah and Villy seemed a bit like Goon mind games to divert attention away from their real intent.
I believe that the Goons intended to build up their supers and titans to be able to roll up to a target region place strategic Keepstars and use their asset superiority to control said region and gradually dominate Eve that way, it is not total dominance, but it is strategic dominance. And it would be a fun thing to do with that massive economic power that they have developed. Of course I could be wrong because this is just thinking about what I would do as a strategic plan and objective based on what the Goons have, but… why not, it is just a pity that the game was not able to handle it…
The feeding of dreads until the MTO Keepstar went down was great spirit.
9-4RPG2 was endurance and bordrom, but not inspireing in any way.
CCP made a video about this, I approve.
Exactly. Changes should be implemented not on server side only, but overall EVE system. Let’s say we have like 50% server TiDi (a strong server is running) and 10000 ships on the grid, then the unimproved EVE clients can cause huge lags and low fps without CUDA engine. Let’s not mix low fps with TiDi (latency). You can have 20% TiDi, but the fps can be above 100. It’s important to rotate camera around smoothly and zoom in and out to ships and other objects.
A boosted server isn’t always a correct solution to improve an application performance. Should be used different mathematical abstraction models and system concepts, like a distributed computation between threads of server and client. For example, as soon on the grid are like over 1000 players the server must open a new thread on a different server and all clients should be redirected there. I don’t know how it works currently, but to open and individual thread for each ship isn’t that bad at all. So, all mentioned operations over ship modules are performed in its own thread. The problem is to synchronize data between all these threads and main server-app thread. D-Scan table must be computed on client side like other UI operations. Local chat must be disabled in that case like other useless extra function in this case.
The application’s performance was never the problem.
The problem was that stuff like controlling fighters, drones, even reloading your guns, they all take server calls. When the server’s too busy to answer, even a 3 way SLI’d 1080 on minimum graphics with all of its rendering done via CUDA will still produce a terrifyingly bad player experience.
What happens when I try to shoot someone on a different server?
I don’t understand the battles and movements of most SoV alliances they did last year. Like TEST operations in Venal or what is the point for NC/PL to help TRI to destroy Fidelas, for example, or like CO2 movement to TRI/FR. It should have been interesting to see that ‘Trioxide’ coalition in the South-East, it’s interesting to see an evolved Fraternity alliance and their coalition. There are many enough powerful groups in nearby Curse, Great Wildlands and Low-Sec, which can join them. Among them, the Legacy was and is the most vulnerably coalition, based on what’s happening around. The TEST is very strong alliance, but the Legacy coalition… IDK.
The Goons leaders looks like much more angry than usual due to what Progodlegend posted just before that glory battle. I suppose, after that Cloud Ring mess they’ll turn to the East, as you mentioned. They have a great support in Stain and that Sax Keepstar in the middle can be used like a staging point and a local bridge beacon. Moreover, their presence in Curse is high as NC, like their allies The Initiative too. It’s hard to predict what the North will do if the Imperium invasion will happen. As I mentioned, the Legacy vulnerability is based on surrounding general situation. The are TEST and BRAVE. The Proviblock just must help Legacy after all, but without a Capital force. The Imperium will not come to take the SoV space, but, possible, to conquer it and give it to their new allies from Stain and all around to create a new CFC block in the South… Who knows…?
Well, I would like to help TEST in this possible war. I’m in a 30 days interval to jump in a Naglfar, Nidhoggur or Hel and, with 4 Large Skill Injectors, I’m in a 100 days interval (until summer) from their Mastery IV. Sadly, at the same time I haven’t enough free time to participate into that Null-Sec ‘scheduled mess’ and I don’t want to buy a PC for $1500, which will hold all that overloaded swarm invasion of ‘space bees’.